{"id":"df239a2d-c327-4227-8097-be366141fc61","entityType":"agent","slug":"clawhub-xiaoba-dev-rotifer-agent","name":"rotifer-agent","canonicalUrl":"https://www.xpersona.co/agent/clawhub-xiaoba-dev-rotifer-agent","canonicalPath":"/agent/clawhub-xiaoba-dev-rotifer-agent","generatedAt":"2026-10-12T18:25:58.434Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T22:22:34.788Z","emptyReason":null},"description":"Build a Rotifer Agent from Genes — decompose intent into capability units, pick Genes by Arena F(g), compose a Rotifer Genome, create the Agent and test it. Use when the user is working with Rotifer specifically: \"Rotifer Agent\", \"compose Genes\", \"Rotifer Genome\", \"Gene composition\", \"rotifer agent create\". Do NOT use for general agent frameworks, or when the words Seq / Par / Cond / Try appear as ordinary programming terms — in Rotifer those name Genome composition strategies, and this Skill applies only when the user is composing Rotifer Genes.","descriptionLabel":"Source description","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 985 downloads reported by the source. Last updated 10/11/2026.","installCommand":"clawhub skill install s174dvd7sg66gw86h4t3c8575d83g51m:rotifer-agent","sourceUrl":"https://clawhub.ai/xiaoba-dev/rotifer-agent","homepage":"https://clawhub.ai/xiaoba-dev/skills/rotifer-agent","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/xiaoba-dev/rotifer-agent","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/xiaoba-dev/skills/rotifer-agent","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":60,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"rotifer-agent technical dossier on Xpersona with agent coverage, OPENCLEW support, and live trust metadata."},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-11T22:22:34.788Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T22:22:34.788Z","emptyReason":null},"stars":null,"forks":null,"downloads":985,"likes":null,"task":null,"library":null,"packageName":null,"latestVersion":"1.2.1","tractionLabel":"985 downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T22:22:34.773Z","emptyReason":null},"lastUpdatedAt":"2026-10-11T22:22:34.788Z","lastCrawledAt":"2026-10-11T22:22:34.773Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-12T22:22:34.773Z","lastVerifiedAt":null,"highlights":[{"version":"1.2.1","createdAt":"2026-08-17T14:53:59.199Z","changelog":"rotifer-agent 1.2.1 - Updated documentation and metadata, including changes to README.md and clawhub.json. - No functional changes to the skill logic or behavior. - Improvements focus on documentation clarity and workflow instructions.","fileCount":5,"zipByteSize":8695},{"version":"1.2.0","createdAt":"2026-08-17T14:38:36.926Z","changelog":"rotifer-agent 1.2.0 - Reworded the skill description for improved clarity and Rotifer-specific focus. - Tightened activation criteria to avoid triggering on unrelated use of terms like Seq/Par. - Removed outdated or redundant documentation files (skill-card.md). - Minor edits for conciseness; core guidance and workflows unchanged.","fileCount":5,"zipByteSize":8198},{"version":"1.1.1","createdAt":"2026-08-17T09:20:22.102Z","changelog":"rotifer-agent 1.1.1 - SKILL.md reformatted; no user-facing behavior changes. - Metadata and documentation files updated for clarity and consistency. - No functional or command changes introduced.","fileCount":5,"zipByteSize":8236},{"version":"1.1.0","createdAt":"2026-08-17T09:10:39.714Z","changelog":"rotifer-agent 1.1.0 - Enhanced CLI guidance and troubleshooting: added `rotifer doctor` and gene-level `rotifer run` steps for easier debugging and workspace validation. - Improved gene selection phase: clarified the difference between local, ranked, and cloud Genes with new commands and explanatory notes. - Updated agent listing and verification: introduced `rotifer agent list` for confirming created agents and genome composition. - Refined troubleshooting in optimization phase: more actionable strategies, including isolated gene testing and broadening gene search. - Documentation cleanup: removed deprecated file and expanded practical tips for CLI usage throughout.","fileCount":5,"zipByteSize":7723},{"version":"1.0.3","createdAt":"2026-04-15T15:13:04.118Z","changelog":"Add Prerequisites section with CLI and MCP setup instructions (no hardcoded versions)","fileCount":5,"zipByteSize":7114},{"version":"1.0.2","createdAt":"2026-04-15T14:02:46.520Z","changelog":"Fix untagged code blocks rendering as invisible text on ClawHub","fileCount":4,"zipByteSize":5788},{"version":"1.0.1","createdAt":"2026-04-15T13:59:34.356Z","changelog":"Translate to English; fix rendering","fileCount":4,"zipByteSize":5787},{"version":"1.0.0","createdAt":"2026-04-15T13:53:14.798Z","changelog":"Initial release: 7-phase Agent architecture from intent to running pipeline","fileCount":4,"zipByteSize":5857}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s174dvd7sg66gw86h4t3c8575d83g51m:rotifer-agent","setupComplexity":"low","setupSteps":["Install using `clawhub skill install s174dvd7sg66gw86h4t3c8575d83g51m:rotifer-agent` in an isolated environment before connecting it to live workloads.","No published capability contract is available yet, so validate auth and request/response behavior manually.","Review the upstream CLAWHUB listing at https://clawhub.ai/xiaoba-dev/rotifer-agent before using production credentials."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xiaoba-dev-rotifer-agent/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xiaoba-dev-rotifer-agent/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xiaoba-dev-rotifer-agent/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-xiaoba-dev-rotifer-agent/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-xiaoba-dev-rotifer-agent/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-xiaoba-dev-rotifer-agent/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-12T18:25:58.431Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xiaoba-dev-rotifer-agent/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xiaoba-dev-rotifer-agent/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xiaoba-dev-rotifer-agent/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-xiaoba-dev-rotifer-agent/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T22:22:34.788Z","emptyReason":null},"readme":"Skill: rotifer-agent\n\nOwner: xiaoba-dev\n\nSummary: Build a Rotifer Agent from Genes — decompose intent into capability units, pick Genes by Arena F(g), compose a Rotifer Genome, create the Agent and test it. Use when the user is working with Rotifer specifically: \"Rotifer Agent\", \"compose Genes\", \"Rotifer Genome\", \"Gene composition\", \"rotifer agent create\". Do NOT use for general agent frameworks, or when the words Seq / Par / Cond / Try appear as ordinary programming terms — in Rotifer those name Genome composition strategies, and this Skill applies only when the user is composing Rotifer Genes.\n\nTags: latest:1.2.1\n\nVersion history:\n\nv1.2.1 | 2026-08-17T14:53:59.199Z | auto\n\nrotifer-agent 1.2.1\n\n- Updated documentation and metadata, including changes to README.md and clawhub.json.\n- No functional changes to the skill logic or behavior.\n- Improvements focus on documentation clarity and workflow instructions.\n\nv1.2.0 | 2026-08-17T14:38:36.926Z | auto\n\nrotifer-agent 1.2.0\n\n- Reworded the skill description for improved clarity and Rotifer-specific focus.\n- Tightened activation criteria to avoid triggering on unrelated use of terms like Seq/Par.\n- Removed outdated or redundant documentation files (skill-card.md).\n- Minor edits for conciseness; core guidance and workflows unchanged.\n\nv1.1.1 | 2026-08-17T09:20:22.102Z | auto\n\nrotifer-agent 1.1.1\n\n- SKILL.md reformatted; no user-facing behavior changes.\n- Metadata and documentation files updated for clarity and consistency.\n- No functional or command changes introduced.\n\nv1.1.0 | 2026-08-17T09:10:39.714Z | auto\n\nrotifer-agent 1.1.0\n\n- Enhanced CLI guidance and troubleshooting: added `rotifer doctor` and gene-level `rotifer run` steps for easier debugging and workspace validation.\n- Improved gene selection phase: clarified the difference between local, ranked, and cloud Genes with new commands and explanatory notes.\n- Updated agent listing and verification: introduced `rotifer agent list` for confirming created agents and genome composition.\n- Refined troubleshooting in optimization phase: more actionable strategies, including isolated gene testing and broadening gene search.\n- Documentation cleanup: removed deprecated file and expanded practical tips for CLI usage throughout.\n\nv1.0.3 | 2026-04-15T15:13:04.118Z | user\n\nAdd Prerequisites section with CLI and MCP setup instructions (no hardcoded versions)\n\nv1.0.2 | 2026-04-15T14:02:46.520Z | user\n\nFix untagged code blocks rendering as invisible text on ClawHub\n\nv1.0.1 | 2026-04-15T13:59:34.356Z | user\n\nTranslate to English; fix rendering\n\nv1.0.0 | 2026-04-15T13:53:14.798Z | user\n\nInitial release: 7-phase Agent architecture from intent to running pipeline\n\nArchive index:\n\nArchive v1.2.1: 5 files, 8695 bytes\n\nFiles: clawhub.json (909b), README.md (3651b), skill-card.md (2197b), SKILL.md (11992b), _meta.json (132b)\n\nFile v1.2.1:SKILL.md\n\n---\nname: rotifer-agent\ndescription: >-\n Build a Rotifer Agent from Genes — decompose intent into capability units, pick Genes by\n Arena F(g), compose a Rotifer Genome, create the Agent and test it.\n Use when the user is working with Rotifer specifically: \"Rotifer Agent\", \"compose Genes\",\n \"Rotifer Genome\", \"Gene composition\", \"rotifer agent create\".\n Do NOT use for general agent frameworks, or when the words Seq / Par / Cond / Try appear as\n ordinary programming terms — in Rotifer those name Genome composition strategies, and this\n Skill applies only when the user is composing Rotifer Genes.\n---\n\n# Rotifer Agent — From Genes to Agents\n\nDecompose user intent into capability units, select Genes from the ecosystem, compose a Genome, create and validate an Agent.\n\n## Prerequisites\n\nThis Skill requires the Rotifer CLI:\n\n```bash\nnpx @rotifer/playground --version\nrotifer doctor\n```\n\n`rotifer doctor` checks the TypeScript→WASM toolchain. Composing an Agent from\nNative Genes means compiling them, and without esbuild and javy that step fails\nwith an error that reads like a code problem.\n\nOr use the MCP Server for IDE integration:\n\n```json\n{\n \"mcpServers\": {\n \"rotifer\": {\n \"command\": \"npx\",\n \"args\": [\"@rotifer/mcp-server\"]\n }\n }\n}\n```\n\n---\n\n**Hierarchy**: Gene (atomic logic) → Genome (composition) → Agent (runnable entity)\n\n---\n\n## Phase 1: Intent Decomposition\n\nBreak the user's goal into independent capability units (each maps to a Gene).\n\n**Steps**:\n\n1. Confirm the Agent's input and expected output with the user\n2. Decompose the task into 2–6 capability units, each satisfying the Gene three axioms (functional cohesion, self-sufficient interface, independently evaluable)\n3. Label each unit with a domain (e.g. `content.grammar`, `security.audit`)\n4. Confirm the decomposition with the user before proceeding to Phase 2\n\n**Output format**:\n\n| # | Capability unit | Domain | Input | Output |\n|---|----------------|--------|-------|--------|\n| 1 | Grammar check | content.grammar | text | issues[], score |\n| 2 | Readability analysis | content.readability | text | grade, suggestions[] |\n\n---\n\n## Phase 2: Gene Selection\n\nMatch existing Genes to each capability unit.\n\n```bash\nrotifer list # what is already in this workspace\nrotifer arena list --domain # ranked by F(g) within one domain\nrotifer search --domain # the Cloud registry — everyone else's Genes\n```\n\nThe three answer different questions, and skipping the third is how a Genome\nends up built only from what happened to be installed already.\n\n**Selection priority**:\n\n| Priority | Source | Command |\n|----------|--------|---------|\n| 1 | Local Gene with highest Arena rank | `rotifer arena list --domain ` |\n| 2 | Cloud Registry | `rotifer search ` → `rotifer info ` → `rotifer install ` |\n| 3 | Doesn't exist, needs creation | Proceed to Phase 3 |\n\nBefore committing a candidate to the Genome, run it on its own — a Gene that\nfails alone will fail inside a pipeline, where the error is much harder to\nlocate:\n\n```bash\nrotifer run --input '{\"...\": \"...\"}'\n```\n\nShow the user candidate Genes' F(g) fitness and fidelity, let them confirm the selection.\n\n---\n\n## Phase 3: Gap Filling\n\nIf a capability unit has no existing Gene:\n\n| Approach | When to use | Action |\n|----------|------------|--------|\n| Create Wrapped Gene | External API / Skill available to wrap | Route to `gene` Skill (dev module) |\n| Create Native Gene | Pure computation, no external dependencies | Route to `gene` Skill (dev module) |\n| Adjust decomposition | Capability unit granularity is wrong | Return to Phase 1 |\n| Merge units | Two units are too coupled, splitting makes the interface awkward | Merge into one Gene |\n\nAfter all Genes are ready, proceed to Phase 4.\n\n---\n\n## Phase 4: Genome Composition\n\nChoose a composition strategy based on relationships between capability units.\n\n### Composition Strategy Decision Table\n\n| Strategy | Semantics | Use when | Example |\n|----------|-----------|----------|---------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Previous output feeds the next | Check → Fix → Format |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results | Grammar check + Readability analysis |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input characteristics determine path | Chinese → Chinese proofing / English → English proofing |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable | Main API → Backup API |\n| **TryPool(A, B, C)** | Race: all try, first success wins | Multiple equivalent implementations | Multiple translation services racing |\n\n### Par Merge Strategies\n\nWhen using Par, specify `--par-merge`:\n\n| Strategy | Behavior | Use when |\n|----------|----------|----------|\n| `first` | Take the first completed result | Racing scenario |\n| `concat` | Concatenate all results (array) | Results are complementary |\n| `merge` | Deep-merge objects | Same structure, merge fields |\n\n### Seq Schema Compatibility Warning\n\n> **Known limitation**: Seq composition requires the previous Gene's outputSchema to be compatible with the next Gene's inputSchema. The current version does not auto-validate — schema mismatches cause runtime errors.\n>\n> **Recommendation**: Before creating a Seq composition, manually compare adjacent Genes' inputSchema / outputSchema in phenotype.json to confirm field names and types match.\n\n### Nested Composition\n\nStrategies can be nested:\n\n```text\nSeq(\n Par(grammar-checker, readability-analyzer),\n tone-analyzer\n)\n```\n\nCorresponding CLI:\n\n```bash\nrotifer agent create doc-qa \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Seq\n```\n\n> The current CLI only supports top-level composition strategies. Nested compositions require manual editing of `.rotifer/agents/.json`.\n\n---\n\n## Phase 5: Agent Creation\n\nExecute creation after confirming the composition plan.\n\n### Manual Gene Selection\n\n```bash\nrotifer agent create \\\n --genes \\\n --composition \\\n --par-merge \n```\n\n### Auto-select Genes (by domain ranking)\n\n```bash\nrotifer agent create \\\n --domain \\\n --top \\\n --composition \n```\n\nAfter creation, verify the Agent configuration file `.rotifer/agents/.json` is correct.\n\n---\n\n## Phase 6: Test Run\n\n```bash\nrotifer agent list\nrotifer agent run --input '{\"text\": \"Test input content\"}'\n```\n\n`rotifer agent list` shows every Agent in the workspace with its state and\ngenome — use it to confirm the Agent was created with the Genes you intended\nbefore running it, and to recover the exact name when a run reports \"agent not\nfound\".\n\n**Validation checklist**:\n\n- Does the output structure match the expected schema?\n- Were all Genes executed? (check logs)\n- Is schema passing correct in Seq composition?\n- Are Par merge results complete?\n- Do error paths (Try/TryPool) degrade correctly?\n\nIf results are unsatisfactory, proceed to Phase 7.\n\n---\n\n## Phase 7: Iterative Optimization\n\n| Problem | Optimization |\n|---------|-------------|\n| One Gene's output quality is poor | `rotifer arena list --domain ` for ranked local alternatives, or `rotifer search ` to look beyond what is installed |\n| Not sure which Gene in the pipeline is at fault | `rotifer run --input '{...}'` on each one in isolation |\n| Seq intermediate results missing fields | Check schema compatibility, consider inserting an adapter Gene |\n| Par merge results are messy | Switch `--par-merge` strategy |\n| Latency too high | Seq → Par (if Genes are independent) |\n| Overall below expectations | Route to `rotifer-arena` Skill for head-to-head Gene evaluation |\n\n---\n\n## Scenario Examples\n\n### Scenario 1: Document Quality Agent\n\n**Goal**: Input text, output grammar issues + readability score + tone analysis.\n\n**Decomposition**:\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Grammar check | grammar-checker | content.grammar |\n| 2 | Readability analysis | readability-analyzer | content.readability |\n| 3 | Tone analysis | tone-analyzer | content.tone |\n\n**Composition**: All three accept `text` input, no dependencies → Par + concat.\n\n```bash\nrotifer agent create doc-quality \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Par \\\n --par-merge concat\n\nrotifer agent run doc-quality --input '{\"text\": \"Document content to check...\"}'\n```\n\n### Scenario 2: Code Review Agent\n\n**Goal**: Input code file, output security vulnerabilities + complexity report + documentation suggestions.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Security audit | security-auditor | security.audit |\n| 2 | Complexity analysis | code-complexity | code.analysis |\n| 3 | Documentation generation | docs-writer | content.docs |\n\n**Composition**: Security audit and complexity analysis can run in parallel, documentation depends on both → Seq(Par(1,2), 3).\n\n```bash\nrotifer agent create code-review \\\n --genes security-auditor code-complexity docs-writer \\\n --composition Seq\n\nrotifer agent run code-review --input '{\"code\": \"...\", \"language\": \"typescript\"}'\n```\n\n> Note: The Par(security-auditor, code-complexity) merged output must be compatible with docs-writer's inputSchema. Manual verification required.\n\n### Scenario 3: Search & Summarize Agent\n\n**Goal**: Input a search query, search → summarize → format output.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Web search | genesis-web-search | search.web |\n| 2 | Text summarization | text-summarizer | content.summarize |\n| 3 | Markdown formatting | markdown-formatter | content.format |\n\n**Composition**: Strict serial pipeline → Seq.\n\n```bash\nrotifer agent create search-digest \\\n --genes genesis-web-search text-summarizer markdown-formatter \\\n --composition Seq\n\nrotifer agent run search-digest --input '{\"query\": \"Rotifer Protocol agent framework\"}'\n```\n\n> Note the Seq schema chain: genesis-web-search output field names must match text-summarizer's inputSchema. Run `cat genes/*/phenotype.json | jq '.inputSchema, .outputSchema'` to verify before creating.\n\n---\n\n## What this Skill does on your machine\n\nIt has no code of its own — it tells your assistant which `rotifer` commands to\nrun. That is why its manifest declares process execution, filesystem read/write\nand outbound network access: every one of those is the CLI acting, not this\nSkill.\n\n| | |\n|---|---|\n| **Runs** | The `rotifer` CLI (`@rotifer/playground`), fetched from npm if not installed. |\n| **Reads** | Genes and Agent definitions in the current project workspace. |\n| **Writes** | Only what the commands below write — Genes into the project's `genes/`, Agent definitions into `.rotifer/agents/`. Nothing outside the project. |\n| **Sends** | Cloud registry and Arena queries, to the public Rotifer API. Your code is not uploaded unless you run `rotifer publish` yourself. |\n\nCommands that install, publish or overwrite are proposed for your approval\nfirst, never run silently.\n\n---\n\n## Related Skills\n\n| Skill | Relationship | When to route |\n|-------|-------------|---------------|\n| `gene` (dev module) | Gene creation/development | Phase 3 gap filling |\n| `rotifer-arena` | Gene comparison & evaluation | Phase 7 when replacing underperforming Genes |\n| `genome` | Genome quality analysis | After Agent creation for overall assessment |\n\n---\n\n## Constraints\n\n- Agent configuration files are stored in `.rotifer/agents/.json` and should not be committed to Git\n- A single Agent should contain 2–6 Genes; more than 6 suggests splitting into multiple Agents\n- Seq schema compatibility is a known limitation — always verify manually before creating\n- Nested compositions require manual JSON editing; the CLI only supports top-level strategies\n\nFile v1.2.1:README.md\n\n# Rotifer Agent\n\n> From intent to running Agent in seven steps.\n\nBuild a Rotifer Agent by composing [Genes](https://rotifer.dev/docs) — the atomic, fitness-evaluable capability units of [Rotifer Protocol](https://rotifer.dev).\n\n**Not a general agent framework.** This composes Rotifer Genes into a Rotifer\nGenome, through the Rotifer CLI. It does not build agents for LangChain, CrewAI\nor anything else, and the words Seq / Par / Cond / Try name Genome composition\nstrategies here — seeing them in ordinary code is not a reason to invoke it.\n\n## What It Does\n\nSay what the Agent should do **in Rotifer terms**, and this Skill handles the architecture:\n\n| Say this | It does this |\n|----------|-------------|\n| \"Build a Rotifer Agent that checks document quality\" | Decomposes into grammar + readability + tone Genes, recommends Par composition |\n| \"Compose Rotifer Genes into a code review Agent\" | Selects security-auditor + code-complexity + docs-writer, designs a Seq(Par, …) Genome |\n| \"Make a Rotifer Genome for search and summarize\" | Picks web-search → summarizer → formatter, wires up Seq |\n\n## The Seven Phases\n\n1. **Intent decomposition** — break the goal into 2-6 capability units\n2. **Gene selection** — find the fittest Gene for each unit from Arena rankings\n3. **Gap filling** — create missing Genes (route to gene → dev module)\n4. **Genome composition** — choose Seq / Par / Cond / Try / TryPool\n5. **Agent creation** — `rotifer agent create`\n6. **Test run** — `rotifer agent run` with validation checklist\n7. **Iteration** — swap underperforming Genes, tune composition\n\n## Composition Strategies\n\n| Strategy | Semantics | Use when |\n|----------|-----------|----------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Output of one feeds the next |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input determines the path |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable |\n| **TryPool(A, B, C)** | Race: all try, first wins | Multiple equivalent implementations |\n\n## Quick Start\n\n```\nBuild a Rotifer Agent that checks document quality — grammar, readability, and tone\n```\n\nThe Skill decomposes the intent, finds top-ranked Genes, and creates the Agent.\n\n## What this Skill does on your machine\n\nIt has no code of its own — it tells your assistant which `rotifer` commands to\nrun. That is what its manifest's process-execution, filesystem and network\npermissions are for: the CLI acting, not this Skill.\n\n| | |\n|---|---|\n| **Runs** | The `rotifer` CLI (`@rotifer/playground`), fetched from npm if not installed. |\n| **Reads** | Genes and Agent definitions in the current project. |\n| **Writes** | Genes into the project's `genes/`, Agent definitions into `.rotifer/agents/`. Nothing outside the project. |\n| **Sends** | Registry and Arena queries to the public Rotifer API. Your code is not uploaded unless you run `rotifer publish` yourself. |\n\nCommands that install, publish or overwrite are proposed for your approval\nfirst, never run silently. Full detail in [SKILL.md](SKILL.md).\n\n## Related Skills\n\n- [rotifer-guide](https://clawhub.ai/skills/rotifer-guide) — Learn Rotifer from scratch\n- [rotifer-arena](https://clawhub.ai/skills/rotifer-arena) — Compare and evaluate Genes\n- [rotifer-self-evolving-agent](https://clawhub.ai/skills/rotifer-self-evolving-agent) — Auto-evolve your Agent\n\n## Links\n\n- [Rotifer Protocol](https://rotifer.dev)\n- [Documentation](https://rotifer.dev/docs)\n- [Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n\n## License\n\nApache-2.0\n\nFile v1.2.1:_meta.json\n\n{\n \"ownerId\": \"kn7ec5m4jmk1h4jyzfrxaesjk9838ykv\",\n \"slug\": \"rotifer-agent\",\n \"version\": \"1.2.1\",\n \"publishedAt\": 1786978439199\n}\n\nFile v1.2.1:skill-card.md\n\n## Description:\n\nBuild a Rotifer Agent from Genes by decomposing intent into capability units, selecting Genes by Arena fitness, composing a Rotifer Genome, creating the Agent, and testing it.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xiaoba-dev](https://clawhub.ai/user/xiaoba-dev)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and engineers use this skill to design Rotifer-specific agents by decomposing goals into Genes, selecting or creating the required Genes, choosing a Genome composition strategy, and validating the resulting Agent.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Unpinned npm execution may run a newer @rotifer/playground or @rotifer/mcp-server version than the reviewer expected.\n\nMitigation: Review or pin the npm package versions before first use.\n\nRisk: Rotifer CLI and MCP commands may read or write project-local Genes and Agent definitions.\n\nMitigation: Run the skill only in workspaces and with network access appropriate for the project data involved.\n\n## Reference(s):\n\n- [Rotifer Protocol](https://rotifer.dev)\n- [Rotifer Documentation](https://rotifer.dev/docs)\n- [Rotifer Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n- [Rotifer Playground](https://github.com/rotifer-protocol/rotifer-playground)\n- [ClawHub Skill Page](https://clawhub.ai/xiaoba-dev/skills/rotifer-agent)\n- [Publisher Profile](https://clawhub.ai/user/xiaoba-dev)\n\n## Skill Output:\n\n**Output Type(s):** [Guidance, Markdown, Shell commands, Configuration]\n\n**Output Format:** [Markdown with tables, checklists, JSON snippets, and inline bash commands]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May propose Rotifer CLI and MCP commands that read and write project-local Rotifer Genes and Agent definitions.]\n\n## Skill Version(s):\n\n1.2.1 (source: server release metadata and clawhub.json)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v1.2.1:clawhub.json\n\n{\n \"name\": \"rotifer-agent\",\n \"display_name\": \"Rotifer Agent\",\n \"description\": \"Build AI Agents from Genes — decompose intent into capability units, select the fittest Genes from Arena rankings, compose them into a Rotifer Genome with Seq/Par/Cond/Try/TryPool strategies, and validate the pipeline end-to-end.\",\n \"tagline\": \"From intent to running Agent in seven steps\",\n \"version\": \"1.2.1\",\n \"author\": \"rotifer-protocol\",\n \"license\": \"Apache-2.0\",\n \"category\": \"developer-tools\",\n \"tags\": [\n \"rotifer\",\n \"ai-agent\",\n \"agent-builder\",\n \"genome\",\n \"composition\",\n \"pipeline\",\n \"gene\",\n \"orchestration\"\n ],\n \"homepage\": \"https://rotifer.dev\",\n \"repository\": \"https://github.com/rotifer-protocol/rotifer-playground\",\n \"permissions\": [\n \"process:exec\",\n \"filesystem:read\",\n \"filesystem:write\",\n \"network:outbound\"\n ],\n \"screenshots\": [],\n \"demo_video\": null\n}\n\nArchive v1.2.0: 5 files, 8198 bytes\n\nFiles: clawhub.json (909b), README.md (2425b), skill-card.md (2325b), SKILL.md (11992b), _meta.json (132b)\n\nFile v1.2.0:SKILL.md\n\n---\nname: rotifer-agent\ndescription: >-\n Build a Rotifer Agent from Genes — decompose intent into capability units, pick Genes by\n Arena F(g), compose a Rotifer Genome, create the Agent and test it.\n Use when the user is working with Rotifer specifically: \"Rotifer Agent\", \"compose Genes\",\n \"Rotifer Genome\", \"Gene composition\", \"rotifer agent create\".\n Do NOT use for general agent frameworks, or when the words Seq / Par / Cond / Try appear as\n ordinary programming terms — in Rotifer those name Genome composition strategies, and this\n Skill applies only when the user is composing Rotifer Genes.\n---\n\n# Rotifer Agent — From Genes to Agents\n\nDecompose user intent into capability units, select Genes from the ecosystem, compose a Genome, create and validate an Agent.\n\n## Prerequisites\n\nThis Skill requires the Rotifer CLI:\n\n```bash\nnpx @rotifer/playground --version\nrotifer doctor\n```\n\n`rotifer doctor` checks the TypeScript→WASM toolchain. Composing an Agent from\nNative Genes means compiling them, and without esbuild and javy that step fails\nwith an error that reads like a code problem.\n\nOr use the MCP Server for IDE integration:\n\n```json\n{\n \"mcpServers\": {\n \"rotifer\": {\n \"command\": \"npx\",\n \"args\": [\"@rotifer/mcp-server\"]\n }\n }\n}\n```\n\n---\n\n**Hierarchy**: Gene (atomic logic) → Genome (composition) → Agent (runnable entity)\n\n---\n\n## Phase 1: Intent Decomposition\n\nBreak the user's goal into independent capability units (each maps to a Gene).\n\n**Steps**:\n\n1. Confirm the Agent's input and expected output with the user\n2. Decompose the task into 2–6 capability units, each satisfying the Gene three axioms (functional cohesion, self-sufficient interface, independently evaluable)\n3. Label each unit with a domain (e.g. `content.grammar`, `security.audit`)\n4. Confirm the decomposition with the user before proceeding to Phase 2\n\n**Output format**:\n\n| # | Capability unit | Domain | Input | Output |\n|---|----------------|--------|-------|--------|\n| 1 | Grammar check | content.grammar | text | issues[], score |\n| 2 | Readability analysis | content.readability | text | grade, suggestions[] |\n\n---\n\n## Phase 2: Gene Selection\n\nMatch existing Genes to each capability unit.\n\n```bash\nrotifer list # what is already in this workspace\nrotifer arena list --domain # ranked by F(g) within one domain\nrotifer search --domain # the Cloud registry — everyone else's Genes\n```\n\nThe three answer different questions, and skipping the third is how a Genome\nends up built only from what happened to be installed already.\n\n**Selection priority**:\n\n| Priority | Source | Command |\n|----------|--------|---------|\n| 1 | Local Gene with highest Arena rank | `rotifer arena list --domain ` |\n| 2 | Cloud Registry | `rotifer search ` → `rotifer info ` → `rotifer install ` |\n| 3 | Doesn't exist, needs creation | Proceed to Phase 3 |\n\nBefore committing a candidate to the Genome, run it on its own — a Gene that\nfails alone will fail inside a pipeline, where the error is much harder to\nlocate:\n\n```bash\nrotifer run --input '{\"...\": \"...\"}'\n```\n\nShow the user candidate Genes' F(g) fitness and fidelity, let them confirm the selection.\n\n---\n\n## Phase 3: Gap Filling\n\nIf a capability unit has no existing Gene:\n\n| Approach | When to use | Action |\n|----------|------------|--------|\n| Create Wrapped Gene | External API / Skill available to wrap | Route to `gene` Skill (dev module) |\n| Create Native Gene | Pure computation, no external dependencies | Route to `gene` Skill (dev module) |\n| Adjust decomposition | Capability unit granularity is wrong | Return to Phase 1 |\n| Merge units | Two units are too coupled, splitting makes the interface awkward | Merge into one Gene |\n\nAfter all Genes are ready, proceed to Phase 4.\n\n---\n\n## Phase 4: Genome Composition\n\nChoose a composition strategy based on relationships between capability units.\n\n### Composition Strategy Decision Table\n\n| Strategy | Semantics | Use when | Example |\n|----------|-----------|----------|---------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Previous output feeds the next | Check → Fix → Format |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results | Grammar check + Readability analysis |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input characteristics determine path | Chinese → Chinese proofing / English → English proofing |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable | Main API → Backup API |\n| **TryPool(A, B, C)** | Race: all try, first success wins | Multiple equivalent implementations | Multiple translation services racing |\n\n### Par Merge Strategies\n\nWhen using Par, specify `--par-merge`:\n\n| Strategy | Behavior | Use when |\n|----------|----------|----------|\n| `first` | Take the first completed result | Racing scenario |\n| `concat` | Concatenate all results (array) | Results are complementary |\n| `merge` | Deep-merge objects | Same structure, merge fields |\n\n### Seq Schema Compatibility Warning\n\n> **Known limitation**: Seq composition requires the previous Gene's outputSchema to be compatible with the next Gene's inputSchema. The current version does not auto-validate — schema mismatches cause runtime errors.\n>\n> **Recommendation**: Before creating a Seq composition, manually compare adjacent Genes' inputSchema / outputSchema in phenotype.json to confirm field names and types match.\n\n### Nested Composition\n\nStrategies can be nested:\n\n```text\nSeq(\n Par(grammar-checker, readability-analyzer),\n tone-analyzer\n)\n```\n\nCorresponding CLI:\n\n```bash\nrotifer agent create doc-qa \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Seq\n```\n\n> The current CLI only supports top-level composition strategies. Nested compositions require manual editing of `.rotifer/agents/.json`.\n\n---\n\n## Phase 5: Agent Creation\n\nExecute creation after confirming the composition plan.\n\n### Manual Gene Selection\n\n```bash\nrotifer agent create \\\n --genes \\\n --composition \\\n --par-merge \n```\n\n### Auto-select Genes (by domain ranking)\n\n```bash\nrotifer agent create \\\n --domain \\\n --top \\\n --composition \n```\n\nAfter creation, verify the Agent configuration file `.rotifer/agents/.json` is correct.\n\n---\n\n## Phase 6: Test Run\n\n```bash\nrotifer agent list\nrotifer agent run --input '{\"text\": \"Test input content\"}'\n```\n\n`rotifer agent list` shows every Agent in the workspace with its state and\ngenome — use it to confirm the Agent was created with the Genes you intended\nbefore running it, and to recover the exact name when a run reports \"agent not\nfound\".\n\n**Validation checklist**:\n\n- Does the output structure match the expected schema?\n- Were all Genes executed? (check logs)\n- Is schema passing correct in Seq composition?\n- Are Par merge results complete?\n- Do error paths (Try/TryPool) degrade correctly?\n\nIf results are unsatisfactory, proceed to Phase 7.\n\n---\n\n## Phase 7: Iterative Optimization\n\n| Problem | Optimization |\n|---------|-------------|\n| One Gene's output quality is poor | `rotifer arena list --domain ` for ranked local alternatives, or `rotifer search ` to look beyond what is installed |\n| Not sure which Gene in the pipeline is at fault | `rotifer run --input '{...}'` on each one in isolation |\n| Seq intermediate results missing fields | Check schema compatibility, consider inserting an adapter Gene |\n| Par merge results are messy | Switch `--par-merge` strategy |\n| Latency too high | Seq → Par (if Genes are independent) |\n| Overall below expectations | Route to `rotifer-arena` Skill for head-to-head Gene evaluation |\n\n---\n\n## Scenario Examples\n\n### Scenario 1: Document Quality Agent\n\n**Goal**: Input text, output grammar issues + readability score + tone analysis.\n\n**Decomposition**:\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Grammar check | grammar-checker | content.grammar |\n| 2 | Readability analysis | readability-analyzer | content.readability |\n| 3 | Tone analysis | tone-analyzer | content.tone |\n\n**Composition**: All three accept `text` input, no dependencies → Par + concat.\n\n```bash\nrotifer agent create doc-quality \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Par \\\n --par-merge concat\n\nrotifer agent run doc-quality --input '{\"text\": \"Document content to check...\"}'\n```\n\n### Scenario 2: Code Review Agent\n\n**Goal**: Input code file, output security vulnerabilities + complexity report + documentation suggestions.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Security audit | security-auditor | security.audit |\n| 2 | Complexity analysis | code-complexity | code.analysis |\n| 3 | Documentation generation | docs-writer | content.docs |\n\n**Composition**: Security audit and complexity analysis can run in parallel, documentation depends on both → Seq(Par(1,2), 3).\n\n```bash\nrotifer agent create code-review \\\n --genes security-auditor code-complexity docs-writer \\\n --composition Seq\n\nrotifer agent run code-review --input '{\"code\": \"...\", \"language\": \"typescript\"}'\n```\n\n> Note: The Par(security-auditor, code-complexity) merged output must be compatible with docs-writer's inputSchema. Manual verification required.\n\n### Scenario 3: Search & Summarize Agent\n\n**Goal**: Input a search query, search → summarize → format output.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Web search | genesis-web-search | search.web |\n| 2 | Text summarization | text-summarizer | content.summarize |\n| 3 | Markdown formatting | markdown-formatter | content.format |\n\n**Composition**: Strict serial pipeline → Seq.\n\n```bash\nrotifer agent create search-digest \\\n --genes genesis-web-search text-summarizer markdown-formatter \\\n --composition Seq\n\nrotifer agent run search-digest --input '{\"query\": \"Rotifer Protocol agent framework\"}'\n```\n\n> Note the Seq schema chain: genesis-web-search output field names must match text-summarizer's inputSchema. Run `cat genes/*/phenotype.json | jq '.inputSchema, .outputSchema'` to verify before creating.\n\n---\n\n## What this Skill does on your machine\n\nIt has no code of its own — it tells your assistant which `rotifer` commands to\nrun. That is why its manifest declares process execution, filesystem read/write\nand outbound network access: every one of those is the CLI acting, not this\nSkill.\n\n| | |\n|---|---|\n| **Runs** | The `rotifer` CLI (`@rotifer/playground`), fetched from npm if not installed. |\n| **Reads** | Genes and Agent definitions in the current project workspace. |\n| **Writes** | Only what the commands below write — Genes into the project's `genes/`, Agent definitions into `.rotifer/agents/`. Nothing outside the project. |\n| **Sends** | Cloud registry and Arena queries, to the public Rotifer API. Your code is not uploaded unless you run `rotifer publish` yourself. |\n\nCommands that install, publish or overwrite are proposed for your approval\nfirst, never run silently.\n\n---\n\n## Related Skills\n\n| Skill | Relationship | When to route |\n|-------|-------------|---------------|\n| `gene` (dev module) | Gene creation/development | Phase 3 gap filling |\n| `rotifer-arena` | Gene comparison & evaluation | Phase 7 when replacing underperforming Genes |\n| `genome` | Genome quality analysis | After Agent creation for overall assessment |\n\n---\n\n## Constraints\n\n- Agent configuration files are stored in `.rotifer/agents/.json` and should not be committed to Git\n- A single Agent should contain 2–6 Genes; more than 6 suggests splitting into multiple Agents\n- Seq schema compatibility is a known limitation — always verify manually before creating\n- Nested compositions require manual JSON editing; the CLI only supports top-level strategies\n\nFile v1.2.0:README.md\n\n# Rotifer Agent\n\n> From intent to running Agent in seven steps.\n\nBuild AI Agents by composing [Genes](https://rotifer.dev/docs) — the atomic, fitness-evaluable capability units of [Rotifer Protocol](https://rotifer.dev).\n\n## What It Does\n\nDescribe what your Agent should do, and this Skill handles the architecture:\n\n| Say this | It does this |\n|----------|-------------|\n| \"Build a document quality checker\" | Decomposes into grammar + readability + tone Genes, recommends Par composition |\n| \"Create a code review agent\" | Selects security-auditor + code-complexity + docs-writer, designs Seq(Par, ...) pipeline |\n| \"Make a search-and-summarize pipeline\" | Picks web-search → summarizer → formatter, wires up Seq |\n\n## The Seven Phases\n\n1. **Intent decomposition** — break the goal into 2-6 capability units\n2. **Gene selection** — find the fittest Gene for each unit from Arena rankings\n3. **Gap filling** — create missing Genes (route to gene → dev module)\n4. **Genome composition** — choose Seq / Par / Cond / Try / TryPool\n5. **Agent creation** — `rotifer agent create`\n6. **Test run** — `rotifer agent run` with validation checklist\n7. **Iteration** — swap underperforming Genes, tune composition\n\n## Composition Strategies\n\n| Strategy | Semantics | Use when |\n|----------|-----------|----------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Output of one feeds the next |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input determines the path |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable |\n| **TryPool(A, B, C)** | Race: all try, first wins | Multiple equivalent implementations |\n\n## Quick Start\n\n```\nBuild an agent that checks document quality — grammar, readability, and tone\n```\n\nThe Skill decomposes the intent, finds top-ranked Genes, and creates the Agent.\n\n## Related Skills\n\n- [rotifer-guide](https://clawhub.ai/skills/rotifer-guide) — Learn Rotifer from scratch\n- [rotifer-arena](https://clawhub.ai/skills/rotifer-arena) — Compare and evaluate Genes\n- [rotifer-self-evolving-agent](https://clawhub.ai/skills/rotifer-self-evolving-agent) — Auto-evolve your Agent\n\n## Links\n\n- [Rotifer Protocol](https://rotifer.dev)\n- [Documentation](https://rotifer.dev/docs)\n- [Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n\n## License\n\nApache-2.0\n\nFile v1.2.0:_meta.json\n\n{\n \"ownerId\": \"kn7ec5m4jmk1h4jyzfrxaesjk9838ykv\",\n \"slug\": \"rotifer-agent\",\n \"version\": \"1.2.0\",\n \"publishedAt\": 1786977516926\n}\n\nFile v1.2.0:skill-card.md\n\n## Description:\n\nBuild a Rotifer Agent from Genes by decomposing intent into capability units, selecting Genes by Arena fitness, composing a Rotifer Genome, creating the Agent, and testing it.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xiaoba-dev](https://clawhub.ai/user/xiaoba-dev)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and engineers use this skill to plan and assemble Rotifer Agents from Genes, including intent decomposition, Gene selection, Genome composition, CLI creation, test runs, and iteration.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Rotifer CLI workflows can read and write project workspace files and contact the Rotifer registry or Arena.\n\nMitigation: Run the skill only in the intended workspace and review proposed npm use, Gene installs, publishes, and overwrites before approval.\n\nRisk: Seq compositions require compatible adjacent Gene schemas, and nested compositions may require manual Agent JSON edits.\n\nMitigation: Compare Gene input and output schemas before creating Seq pipelines and inspect generated Agent configuration before test runs.\n\n## Reference(s):\n\n- [Rotifer Agent on ClawHub](https://clawhub.ai/xiaoba-dev/skills/rotifer-agent)\n- [Rotifer Protocol](https://rotifer.dev)\n- [Rotifer Documentation](https://rotifer.dev/docs)\n- [Rotifer Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n- [Rotifer Playground Repository](https://github.com/rotifer-protocol/rotifer-playground)\n\n## Skill Output:\n\n**Output Type(s):** [Guidance, Markdown, Shell commands, Configuration instructions, JSON]\n\n**Output Format:** [Markdown with tables, checklists, JSON snippets, and bash command blocks]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May propose Rotifer CLI commands for local Agent creation, Gene installation, registry queries, and validation; installs, publishes, and overwrites require user review.]\n\n## Skill Version(s):\n\n1.2.0 (source: server release evidence and clawhub.json)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v1.2.0:clawhub.json\n\n{\n \"name\": \"rotifer-agent\",\n \"display_name\": \"Rotifer Agent\",\n \"description\": \"Build AI Agents from Genes — decompose intent into capability units, select the fittest Genes from Arena rankings, compose them into a Rotifer Genome with Seq/Par/Cond/Try/TryPool strategies, and validate the pipeline end-to-end.\",\n \"tagline\": \"From intent to running Agent in seven steps\",\n \"version\": \"1.2.0\",\n \"author\": \"rotifer-protocol\",\n \"license\": \"Apache-2.0\",\n \"category\": \"developer-tools\",\n \"tags\": [\n \"rotifer\",\n \"ai-agent\",\n \"agent-builder\",\n \"genome\",\n \"composition\",\n \"pipeline\",\n \"gene\",\n \"orchestration\"\n ],\n \"homepage\": \"https://rotifer.dev\",\n \"repository\": \"https://github.com/rotifer-protocol/rotifer-playground\",\n \"permissions\": [\n \"process:exec\",\n \"filesystem:read\",\n \"filesystem:write\",\n \"network:outbound\"\n ],\n \"screenshots\": [],\n \"demo_video\": null\n}\n\nArchive v1.1.1: 5 files, 8236 bytes\n\nFiles: clawhub.json (887b), README.md (2425b), skill-card.md (2506b), SKILL.md (11813b), _meta.json (132b)\n\nFile v1.1.1:SKILL.md\n\n---\nname: rotifer-agent\ndescription: >-\n End-to-end guide for building AI Agents from Genes: intent decomposition, Gene selection,\n Genome composition, Agent creation, and testing.\n Use when the user mentions \"build Agent\", \"compose Genes\", \"Agent composition\", \"agent create\",\n \"agent run\", \"Genome\", \"assemble Agent\", \"Agent architecture\", \"composition strategy\",\n \"Seq\", \"Par\", \"Cond\", \"Try\", \"TryPool\", \"synthesize Agent\".\n---\n\n# Rotifer Agent — From Genes to Agents\n\nDecompose user intent into capability units, select Genes from the ecosystem, compose a Genome, create and validate an Agent.\n\n## Prerequisites\n\nThis Skill requires the Rotifer CLI:\n\n```bash\nnpx @rotifer/playground --version\nrotifer doctor\n```\n\n`rotifer doctor` checks the TypeScript→WASM toolchain. Composing an Agent from\nNative Genes means compiling them, and without esbuild and javy that step fails\nwith an error that reads like a code problem.\n\nOr use the MCP Server for IDE integration:\n\n```json\n{\n \"mcpServers\": {\n \"rotifer\": {\n \"command\": \"npx\",\n \"args\": [\"@rotifer/mcp-server\"]\n }\n }\n}\n```\n\n---\n\n**Hierarchy**: Gene (atomic logic) → Genome (composition) → Agent (runnable entity)\n\n---\n\n## Phase 1: Intent Decomposition\n\nBreak the user's goal into independent capability units (each maps to a Gene).\n\n**Steps**:\n\n1. Confirm the Agent's input and expected output with the user\n2. Decompose the task into 2–6 capability units, each satisfying the Gene three axioms (functional cohesion, self-sufficient interface, independently evaluable)\n3. Label each unit with a domain (e.g. `content.grammar`, `security.audit`)\n4. Confirm the decomposition with the user before proceeding to Phase 2\n\n**Output format**:\n\n| # | Capability unit | Domain | Input | Output |\n|---|----------------|--------|-------|--------|\n| 1 | Grammar check | content.grammar | text | issues[], score |\n| 2 | Readability analysis | content.readability | text | grade, suggestions[] |\n\n---\n\n## Phase 2: Gene Selection\n\nMatch existing Genes to each capability unit.\n\n```bash\nrotifer list # what is already in this workspace\nrotifer arena list --domain # ranked by F(g) within one domain\nrotifer search --domain # the Cloud registry — everyone else's Genes\n```\n\nThe three answer different questions, and skipping the third is how a Genome\nends up built only from what happened to be installed already.\n\n**Selection priority**:\n\n| Priority | Source | Command |\n|----------|--------|---------|\n| 1 | Local Gene with highest Arena rank | `rotifer arena list --domain ` |\n| 2 | Cloud Registry | `rotifer search ` → `rotifer info ` → `rotifer install ` |\n| 3 | Doesn't exist, needs creation | Proceed to Phase 3 |\n\nBefore committing a candidate to the Genome, run it on its own — a Gene that\nfails alone will fail inside a pipeline, where the error is much harder to\nlocate:\n\n```bash\nrotifer run --input '{\"...\": \"...\"}'\n```\n\nShow the user candidate Genes' F(g) fitness and fidelity, let them confirm the selection.\n\n---\n\n## Phase 3: Gap Filling\n\nIf a capability unit has no existing Gene:\n\n| Approach | When to use | Action |\n|----------|------------|--------|\n| Create Wrapped Gene | External API / Skill available to wrap | Route to `gene` Skill (dev module) |\n| Create Native Gene | Pure computation, no external dependencies | Route to `gene` Skill (dev module) |\n| Adjust decomposition | Capability unit granularity is wrong | Return to Phase 1 |\n| Merge units | Two units are too coupled, splitting makes the interface awkward | Merge into one Gene |\n\nAfter all Genes are ready, proceed to Phase 4.\n\n---\n\n## Phase 4: Genome Composition\n\nChoose a composition strategy based on relationships between capability units.\n\n### Composition Strategy Decision Table\n\n| Strategy | Semantics | Use when | Example |\n|----------|-----------|----------|---------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Previous output feeds the next | Check → Fix → Format |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results | Grammar check + Readability analysis |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input characteristics determine path | Chinese → Chinese proofing / English → English proofing |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable | Main API → Backup API |\n| **TryPool(A, B, C)** | Race: all try, first success wins | Multiple equivalent implementations | Multiple translation services racing |\n\n### Par Merge Strategies\n\nWhen using Par, specify `--par-merge`:\n\n| Strategy | Behavior | Use when |\n|----------|----------|----------|\n| `first` | Take the first completed result | Racing scenario |\n| `concat` | Concatenate all results (array) | Results are complementary |\n| `merge` | Deep-merge objects | Same structure, merge fields |\n\n### Seq Schema Compatibility Warning\n\n> **Known limitation**: Seq composition requires the previous Gene's outputSchema to be compatible with the next Gene's inputSchema. The current version does not auto-validate — schema mismatches cause runtime errors.\n>\n> **Recommendation**: Before creating a Seq composition, manually compare adjacent Genes' inputSchema / outputSchema in phenotype.json to confirm field names and types match.\n\n### Nested Composition\n\nStrategies can be nested:\n\n```text\nSeq(\n Par(grammar-checker, readability-analyzer),\n tone-analyzer\n)\n```\n\nCorresponding CLI:\n\n```bash\nrotifer agent create doc-qa \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Seq\n```\n\n> The current CLI only supports top-level composition strategies. Nested compositions require manual editing of `.rotifer/agents/.json`.\n\n---\n\n## Phase 5: Agent Creation\n\nExecute creation after confirming the composition plan.\n\n### Manual Gene Selection\n\n```bash\nrotifer agent create \\\n --genes \\\n --composition \\\n --par-merge \n```\n\n### Auto-select Genes (by domain ranking)\n\n```bash\nrotifer agent create \\\n --domain \\\n --top \\\n --composition \n```\n\nAfter creation, verify the Agent configuration file `.rotifer/agents/.json` is correct.\n\n---\n\n## Phase 6: Test Run\n\n```bash\nrotifer agent list\nrotifer agent run --input '{\"text\": \"Test input content\"}'\n```\n\n`rotifer agent list` shows every Agent in the workspace with its state and\ngenome — use it to confirm the Agent was created with the Genes you intended\nbefore running it, and to recover the exact name when a run reports \"agent not\nfound\".\n\n**Validation checklist**:\n\n- Does the output structure match the expected schema?\n- Were all Genes executed? (check logs)\n- Is schema passing correct in Seq composition?\n- Are Par merge results complete?\n- Do error paths (Try/TryPool) degrade correctly?\n\nIf results are unsatisfactory, proceed to Phase 7.\n\n---\n\n## Phase 7: Iterative Optimization\n\n| Problem | Optimization |\n|---------|-------------|\n| One Gene's output quality is poor | `rotifer arena list --domain ` for ranked local alternatives, or `rotifer search ` to look beyond what is installed |\n| Not sure which Gene in the pipeline is at fault | `rotifer run --input '{...}'` on each one in isolation |\n| Seq intermediate results missing fields | Check schema compatibility, consider inserting an adapter Gene |\n| Par merge results are messy | Switch `--par-merge` strategy |\n| Latency too high | Seq → Par (if Genes are independent) |\n| Overall below expectations | Route to `rotifer-arena` Skill for head-to-head Gene evaluation |\n\n---\n\n## Scenario Examples\n\n### Scenario 1: Document Quality Agent\n\n**Goal**: Input text, output grammar issues + readability score + tone analysis.\n\n**Decomposition**:\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Grammar check | grammar-checker | content.grammar |\n| 2 | Readability analysis | readability-analyzer | content.readability |\n| 3 | Tone analysis | tone-analyzer | content.tone |\n\n**Composition**: All three accept `text` input, no dependencies → Par + concat.\n\n```bash\nrotifer agent create doc-quality \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Par \\\n --par-merge concat\n\nrotifer agent run doc-quality --input '{\"text\": \"Document content to check...\"}'\n```\n\n### Scenario 2: Code Review Agent\n\n**Goal**: Input code file, output security vulnerabilities + complexity report + documentation suggestions.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Security audit | security-auditor | security.audit |\n| 2 | Complexity analysis | code-complexity | code.analysis |\n| 3 | Documentation generation | docs-writer | content.docs |\n\n**Composition**: Security audit and complexity analysis can run in parallel, documentation depends on both → Seq(Par(1,2), 3).\n\n```bash\nrotifer agent create code-review \\\n --genes security-auditor code-complexity docs-writer \\\n --composition Seq\n\nrotifer agent run code-review --input '{\"code\": \"...\", \"language\": \"typescript\"}'\n```\n\n> Note: The Par(security-auditor, code-complexity) merged output must be compatible with docs-writer's inputSchema. Manual verification required.\n\n### Scenario 3: Search & Summarize Agent\n\n**Goal**: Input a search query, search → summarize → format output.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Web search | genesis-web-search | search.web |\n| 2 | Text summarization | text-summarizer | content.summarize |\n| 3 | Markdown formatting | markdown-formatter | content.format |\n\n**Composition**: Strict serial pipeline → Seq.\n\n```bash\nrotifer agent create search-digest \\\n --genes genesis-web-search text-summarizer markdown-formatter \\\n --composition Seq\n\nrotifer agent run search-digest --input '{\"query\": \"Rotifer Protocol agent framework\"}'\n```\n\n> Note the Seq schema chain: genesis-web-search output field names must match text-summarizer's inputSchema. Run `cat genes/*/phenotype.json | jq '.inputSchema, .outputSchema'` to verify before creating.\n\n---\n\n## What this Skill does on your machine\n\nIt has no code of its own — it tells your assistant which `rotifer` commands to\nrun. That is why its manifest declares process execution, filesystem read/write\nand outbound network access: every one of those is the CLI acting, not this\nSkill.\n\n| | |\n|---|---|\n| **Runs** | The `rotifer` CLI (`@rotifer/playground`), fetched from npm if not installed. |\n| **Reads** | Genes and Agent definitions in the current project workspace. |\n| **Writes** | Only what the commands below write — Genes into the project's `genes/`, Agent definitions into `.rotifer/agents/`. Nothing outside the project. |\n| **Sends** | Cloud registry and Arena queries, to the public Rotifer API. Your code is not uploaded unless you run `rotifer publish` yourself. |\n\nCommands that install, publish or overwrite are proposed for your approval\nfirst, never run silently.\n\n---\n\n## Related Skills\n\n| Skill | Relationship | When to route |\n|-------|-------------|---------------|\n| `gene` (dev module) | Gene creation/development | Phase 3 gap filling |\n| `rotifer-arena` | Gene comparison & evaluation | Phase 7 when replacing underperforming Genes |\n| `genome` | Genome quality analysis | After Agent creation for overall assessment |\n\n---\n\n## Constraints\n\n- Agent configuration files are stored in `.rotifer/agents/.json` and should not be committed to Git\n- A single Agent should contain 2–6 Genes; more than 6 suggests splitting into multiple Agents\n- Seq schema compatibility is a known limitation — always verify manually before creating\n- Nested compositions require manual JSON editing; the CLI only supports top-level strategies\n\nFile v1.1.1:README.md\n\n# Rotifer Agent\n\n> From intent to running Agent in seven steps.\n\nBuild AI Agents by composing [Genes](https://rotifer.dev/docs) — the atomic, fitness-evaluable capability units of [Rotifer Protocol](https://rotifer.dev).\n\n## What It Does\n\nDescribe what your Agent should do, and this Skill handles the architecture:\n\n| Say this | It does this |\n|----------|-------------|\n| \"Build a document quality checker\" | Decomposes into grammar + readability + tone Genes, recommends Par composition |\n| \"Create a code review agent\" | Selects security-auditor + code-complexity + docs-writer, designs Seq(Par, ...) pipeline |\n| \"Make a search-and-summarize pipeline\" | Picks web-search → summarizer → formatter, wires up Seq |\n\n## The Seven Phases\n\n1. **Intent decomposition** — break the goal into 2-6 capability units\n2. **Gene selection** — find the fittest Gene for each unit from Arena rankings\n3. **Gap filling** — create missing Genes (route to gene → dev module)\n4. **Genome composition** — choose Seq / Par / Cond / Try / TryPool\n5. **Agent creation** — `rotifer agent create`\n6. **Test run** — `rotifer agent run` with validation checklist\n7. **Iteration** — swap underperforming Genes, tune composition\n\n## Composition Strategies\n\n| Strategy | Semantics | Use when |\n|----------|-----------|----------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Output of one feeds the next |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input determines the path |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable |\n| **TryPool(A, B, C)** | Race: all try, first wins | Multiple equivalent implementations |\n\n## Quick Start\n\n```\nBuild an agent that checks document quality — grammar, readability, and tone\n```\n\nThe Skill decomposes the intent, finds top-ranked Genes, and creates the Agent.\n\n## Related Skills\n\n- [rotifer-guide](https://clawhub.ai/skills/rotifer-guide) — Learn Rotifer from scratch\n- [rotifer-arena](https://clawhub.ai/skills/rotifer-arena) — Compare and evaluate Genes\n- [rotifer-self-evolving-agent](https://clawhub.ai/skills/rotifer-self-evolving-agent) — Auto-evolve your Agent\n\n## Links\n\n- [Rotifer Protocol](https://rotifer.dev)\n- [Documentation](https://rotifer.dev/docs)\n- [Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n\n## License\n\nApache-2.0\n\nFile v1.1.1:_meta.json\n\n{\n \"ownerId\": \"kn7ec5m4jmk1h4jyzfrxaesjk9838ykv\",\n \"slug\": \"rotifer-agent\",\n \"version\": \"1.1.1\",\n \"publishedAt\": 1786958422102\n}\n\nFile v1.1.1:skill-card.md\n\n## Description:\n\nBuild AI agents from Rotifer Genes by decomposing intent, selecting candidate Genes, composing a Genome, creating the Agent, and validating the result.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xiaoba-dev](https://clawhub.ai/user/xiaoba-dev)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and engineers use this skill to turn an agent idea into a Rotifer Agent plan, select Genes, choose Seq/Par/Cond/Try/TryPool composition strategies, create the Agent, and run validation checks.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Rotifer CLI usage may fetch npm packages, inspect local Genes, write Agent definitions, and make registry or Arena network requests.\n\nMitigation: Install and run the skill only when Rotifer tooling is intended, and review Genes, npm packages, and proposed commands before execution.\n\nRisk: Publish commands can expose user-selected project artifacts if run intentionally.\n\nMitigation: Review any publish command and its target before approval; the artifact states code is not uploaded unless publish is run.\n\nRisk: Seq and nested compositions can fail at runtime when adjacent Gene schemas or manually edited Agent definitions do not match.\n\nMitigation: Manually compare input and output schemas before creating Seq compositions, and validate generated Agent configuration before running it.\n\n## Reference(s):\n\n- [Rotifer Agent Skill Page](https://clawhub.ai/xiaoba-dev/skills/rotifer-agent)\n- [Rotifer Protocol](https://rotifer.dev)\n- [Rotifer Documentation](https://rotifer.dev/docs)\n- [Rotifer Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n- [Rotifer Playground Repository](https://github.com/rotifer-protocol/rotifer-playground)\n\n## Skill Output:\n\n**Output Type(s):** [guidance, markdown, shell commands, configuration]\n\n**Output Format:** [Markdown guidance with command snippets, tables, and JSON configuration examples]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Guides Rotifer CLI usage and may propose workspace file changes that should be reviewed before execution.]\n\n## Skill Version(s):\n\n1.1.1 (source: server release metadata and clawhub.json)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v1.1.1:clawhub.json\n\n{\n \"name\": \"rotifer-agent\",\n \"display_name\": \"Rotifer Agent\",\n \"description\": \"Build AI Agents from Genes — decompose intent into capability units, select the fittest Genes from Arena rankings, compose them with Seq/Par/Cond/Try/TryPool strategies, and validate the pipeline end-to-end.\",\n \"tagline\": \"From intent to running Agent in seven steps\",\n \"version\": \"1.1.1\",\n \"author\": \"rotifer-protocol\",\n \"license\": \"Apache-2.0\",\n \"category\": \"developer-tools\",\n \"tags\": [\n \"rotifer\",\n \"ai-agent\",\n \"agent-builder\",\n \"genome\",\n \"composition\",\n \"pipeline\",\n \"gene\",\n \"orchestration\"\n ],\n \"homepage\": \"https://rotifer.dev\",\n \"repository\": \"https://github.com/rotifer-protocol/rotifer-playground\",\n \"permissions\": [\n \"process:exec\",\n \"filesystem:read\",\n \"filesystem:write\",\n \"network:outbound\"\n ],\n \"screenshots\": [],\n \"demo_video\": null\n}\n\nArchive v1.1.0: 5 files, 7723 bytes\n\nFiles: clawhub.json (887b), README.md (2425b), skill-card.md (2243b), SKILL.md (10921b), _meta.json (132b)\n\nFile v1.1.0:SKILL.md\n\n---\nname: rotifer-agent\ndescription: >-\n End-to-end guide for building AI Agents from Genes: intent decomposition, Gene selection,\n Genome composition, Agent creation, and testing.\n Use when the user mentions \"build Agent\", \"compose Genes\", \"Agent composition\", \"agent create\",\n \"agent run\", \"Genome\", \"assemble Agent\", \"Agent architecture\", \"composition strategy\",\n \"Seq\", \"Par\", \"Cond\", \"Try\", \"TryPool\", \"synthesize Agent\".\n---\n\n# Rotifer Agent — From Genes to Agents\n\nDecompose user intent into capability units, select Genes from the ecosystem, compose a Genome, create and validate an Agent.\n\n## Prerequisites\n\nThis Skill requires the Rotifer CLI:\n\n```bash\nnpx @rotifer/playground --version\nrotifer doctor\n```\n\n`rotifer doctor` checks the TypeScript→WASM toolchain. Composing an Agent from\nNative Genes means compiling them, and without esbuild and javy that step fails\nwith an error that reads like a code problem.\n\nOr use the MCP Server for IDE integration:\n\n```json\n{\n \"mcpServers\": {\n \"rotifer\": {\n \"command\": \"npx\",\n \"args\": [\"@rotifer/mcp-server\"]\n }\n }\n}\n```\n\n---\n\n**Hierarchy**: Gene (atomic logic) → Genome (composition) → Agent (runnable entity)\n\n---\n\n## Phase 1: Intent Decomposition\n\nBreak the user's goal into independent capability units (each maps to a Gene).\n\n**Steps**:\n\n1. Confirm the Agent's input and expected output with the user\n2. Decompose the task into 2–6 capability units, each satisfying the Gene three axioms (functional cohesion, self-sufficient interface, independently evaluable)\n3. Label each unit with a domain (e.g. `content.grammar`, `security.audit`)\n4. Confirm the decomposition with the user before proceeding to Phase 2\n\n**Output format**:\n\n| # | Capability unit | Domain | Input | Output |\n|---|----------------|--------|-------|--------|\n| 1 | Grammar check | content.grammar | text | issues[], score |\n| 2 | Readability analysis | content.readability | text | grade, suggestions[] |\n\n---\n\n## Phase 2: Gene Selection\n\nMatch existing Genes to each capability unit.\n\n```bash\nrotifer list # what is already in this workspace\nrotifer arena list --domain # ranked by F(g) within one domain\nrotifer search --domain # the Cloud registry — everyone else's Genes\n```\n\nThe three answer different questions, and skipping the third is how a Genome\nends up built only from what happened to be installed already.\n\n**Selection priority**:\n\n| Priority | Source | Command |\n|----------|--------|---------|\n| 1 | Local Gene with highest Arena rank | `rotifer arena list --domain ` |\n| 2 | Cloud Registry | `rotifer search ` → `rotifer info ` → `rotifer install ` |\n| 3 | Doesn't exist, needs creation | Proceed to Phase 3 |\n\nBefore committing a candidate to the Genome, run it on its own — a Gene that\nfails alone will fail inside a pipeline, where the error is much harder to\nlocate:\n\n```bash\nrotifer run --input '{\"...\": \"...\"}'\n```\n\nShow the user candidate Genes' F(g) fitness and fidelity, let them confirm the selection.\n\n---\n\n## Phase 3: Gap Filling\n\nIf a capability unit has no existing Gene:\n\n| Approach | When to use | Action |\n|----------|------------|--------|\n| Create Wrapped Gene | External API / Skill available to wrap | Route to `gene` Skill (dev module) |\n| Create Native Gene | Pure computation, no external dependencies | Route to `gene` Skill (dev module) |\n| Adjust decomposition | Capability unit granularity is wrong | Return to Phase 1 |\n| Merge units | Two units are too coupled, splitting makes the interface awkward | Merge into one Gene |\n\nAfter all Genes are ready, proceed to Phase 4.\n\n---\n\n## Phase 4: Genome Composition\n\nChoose a composition strategy based on relationships between capability units.\n\n### Composition Strategy Decision Table\n\n| Strategy | Semantics | Use when | Example |\n|----------|-----------|----------|---------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Previous output feeds the next | Check → Fix → Format |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results | Grammar check + Readability analysis |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input characteristics determine path | Chinese → Chinese proofing / English → English proofing |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable | Main API → Backup API |\n| **TryPool(A, B, C)** | Race: all try, first success wins | Multiple equivalent implementations | Multiple translation services racing |\n\n### Par Merge Strategies\n\nWhen using Par, specify `--par-merge`:\n\n| Strategy | Behavior | Use when |\n|----------|----------|----------|\n| `first` | Take the first completed result | Racing scenario |\n| `concat` | Concatenate all results (array) | Results are complementary |\n| `merge` | Deep-merge objects | Same structure, merge fields |\n\n### Seq Schema Compatibility Warning\n\n> **Known limitation**: Seq composition requires the previous Gene's outputSchema to be compatible with the next Gene's inputSchema. The current version does not auto-validate — schema mismatches cause runtime errors.\n>\n> **Recommendation**: Before creating a Seq composition, manually compare adjacent Genes' inputSchema / outputSchema in phenotype.json to confirm field names and types match.\n\n### Nested Composition\n\nStrategies can be nested:\n\n```text\nSeq(\n Par(grammar-checker, readability-analyzer),\n tone-analyzer\n)\n```\n\nCorresponding CLI:\n\n```bash\nrotifer agent create doc-qa \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Seq\n```\n\n> The current CLI only supports top-level composition strategies. Nested compositions require manual editing of `.rotifer/agents/.json`.\n\n---\n\n## Phase 5: Agent Creation\n\nExecute creation after confirming the composition plan.\n\n### Manual Gene Selection\n\n```bash\nrotifer agent create \\\n --genes \\\n --composition \\\n --par-merge \n```\n\n### Auto-select Genes (by domain ranking)\n\n```bash\nrotifer agent create \\\n --domain \\\n --top \\\n --composition \n```\n\nAfter creation, verify the Agent configuration file `.rotifer/agents/.json` is correct.\n\n---\n\n## Phase 6: Test Run\n\n```bash\nrotifer agent list\nrotifer agent run --input '{\"text\": \"Test input content\"}'\n```\n\n`rotifer agent list` shows every Agent in the workspace with its state and\ngenome — use it to confirm the Agent was created with the Genes you intended\nbefore running it, and to recover the exact name when a run reports \"agent not\nfound\".\n\n**Validation checklist**:\n\n- Does the output structure match the expected schema?\n- Were all Genes executed? (check logs)\n- Is schema passing correct in Seq composition?\n- Are Par merge results complete?\n- Do error paths (Try/TryPool) degrade correctly?\n\nIf results are unsatisfactory, proceed to Phase 7.\n\n---\n\n## Phase 7: Iterative Optimization\n\n| Problem | Optimization |\n|---------|-------------|\n| One Gene's output quality is poor | `rotifer arena list --domain ` for ranked local alternatives, or `rotifer search ` to look beyond what is installed |\n| Not sure which Gene in the pipeline is at fault | `rotifer run --input '{...}'` on each one in isolation |\n| Seq intermediate results missing fields | Check schema compatibility, consider inserting an adapter Gene |\n| Par merge results are messy | Switch `--par-merge` strategy |\n| Latency too high | Seq → Par (if Genes are independent) |\n| Overall below expectations | Route to `rotifer-arena` Skill for head-to-head Gene evaluation |\n\n---\n\n## Scenario Examples\n\n### Scenario 1: Document Quality Agent\n\n**Goal**: Input text, output grammar issues + readability score + tone analysis.\n\n**Decomposition**:\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Grammar check | grammar-checker | content.grammar |\n| 2 | Readability analysis | readability-analyzer | content.readability |\n| 3 | Tone analysis | tone-analyzer | content.tone |\n\n**Composition**: All three accept `text` input, no dependencies → Par + concat.\n\n```bash\nrotifer agent create doc-quality \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Par \\\n --par-merge concat\n\nrotifer agent run doc-quality --input '{\"text\": \"Document content to check...\"}'\n```\n\n### Scenario 2: Code Review Agent\n\n**Goal**: Input code file, output security vulnerabilities + complexity report + documentation suggestions.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Security audit | security-auditor | security.audit |\n| 2 | Complexity analysis | code-complexity | code.analysis |\n| 3 | Documentation generation | docs-writer | content.docs |\n\n**Composition**: Security audit and complexity analysis can run in parallel, documentation depends on both → Seq(Par(1,2), 3).\n\n```bash\nrotifer agent create code-review \\\n --genes security-auditor code-complexity docs-writer \\\n --composition Seq\n\nrotifer agent run code-review --input '{\"code\": \"...\", \"language\": \"typescript\"}'\n```\n\n> Note: The Par(security-auditor, code-complexity) merged output must be compatible with docs-writer's inputSchema. Manual verification required.\n\n### Scenario 3: Search & Summarize Agent\n\n**Goal**: Input a search query, search → summarize → format output.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Web search | genesis-web-search | search.web |\n| 2 | Text summarization | text-summarizer | content.summarize |\n| 3 | Markdown formatting | markdown-formatter | content.format |\n\n**Composition**: Strict serial pipeline → Seq.\n\n```bash\nrotifer agent create search-digest \\\n --genes genesis-web-search text-summarizer markdown-formatter \\\n --composition Seq\n\nrotifer agent run search-digest --input '{\"query\": \"Rotifer Protocol agent framework\"}'\n```\n\n> Note the Seq schema chain: genesis-web-search output field names must match text-summarizer's inputSchema. Run `cat genes/*/phenotype.json | jq '.inputSchema, .outputSchema'` to verify before creating.\n\n---\n\n## Related Skills\n\n| Skill | Relationship | When to route |\n|-------|-------------|---------------|\n| `gene` (dev module) | Gene creation/development | Phase 3 gap filling |\n| `rotifer-arena` | Gene comparison & evaluation | Phase 7 when replacing underperforming Genes |\n| `genome` | Genome quality analysis | After Agent creation for overall assessment |\n\n---\n\n## Constraints\n\n- Agent configuration files are stored in `.rotifer/agents/.json` and should not be committed to Git\n- A single Agent should contain 2–6 Genes; more than 6 suggests splitting into multiple Agents\n- Seq schema compatibility is a known limitation — always verify manually before creating\n- Nested compositions require manual JSON editing; the CLI only supports top-level strategies\n\nFile v1.1.0:README.md\n\n# Rotifer Agent\n\n> From intent to running Agent in seven steps.\n\nBuild AI Agents by composing [Genes](https://rotifer.dev/docs) — the atomic, fitness-evaluable capability units of [Rotifer Protocol](https://rotifer.dev).\n\n## What It Does\n\nDescribe what your Agent should do, and this Skill handles the architecture:\n\n| Say this | It does this |\n|----------|-------------|\n| \"Build a document quality checker\" | Decomposes into grammar + readability + tone Genes, recommends Par composition |\n| \"Create a code review agent\" | Selects security-auditor + code-complexity + docs-writer, designs Seq(Par, ...) pipeline |\n| \"Make a search-and-summarize pipeline\" | Picks web-search → summarizer → formatter, wires up Seq |\n\n## The Seven Phases\n\n1. **Intent decomposition** — break the goal into 2-6 capability units\n2. **Gene selection** — find the fittest Gene for each unit from Arena rankings\n3. **Gap filling** — create missing Genes (route to gene → dev module)\n4. **Genome composition** — choose Seq / Par / Cond / Try / TryPool\n5. **Agent creation** — `rotifer agent create`\n6. **Test run** — `rotifer agent run` with validation checklist\n7. **Iteration** — swap underperforming Genes, tune composition\n\n## Composition Strategies\n\n| Strategy | Semantics | Use when |\n|----------|-----------|----------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Output of one feeds the next |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input determines the path |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable |\n| **TryPool(A, B, C)** | Race: all try, first wins | Multiple equivalent implementations |\n\n## Quick Start\n\n```\nBuild an agent that checks document quality — grammar, readability, and tone\n```\n\nThe Skill decomposes the intent, finds top-ranked Genes, and creates the Agent.\n\n## Related Skills\n\n- [rotifer-guide](https://clawhub.ai/skills/rotifer-guide) — Learn Rotifer from scratch\n- [rotifer-arena](https://clawhub.ai/skills/rotifer-arena) — Compare and evaluate Genes\n- [rotifer-self-evolving-agent](https://clawhub.ai/skills/rotifer-self-evolving-agent) — Auto-evolve your Agent\n\n## Links\n\n- [Rotifer Protocol](https://rotifer.dev)\n- [Documentation](https://rotifer.dev/docs)\n- [Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n\n## License\n\nApache-2.0\n\nFile v1.1.0:_meta.json\n\n{\n \"ownerId\": \"kn7ec5m4jmk1h4jyzfrxaesjk9838ykv\",\n \"slug\": \"rotifer-agent\",\n \"version\": \"1.1.0\",\n \"publishedAt\": 1786957839714\n}\n\nFile v1.1.0:skill-card.md\n\n## Description:\n\nGuides developers through building AI Agents from Rotifer Genes by decomposing intent, selecting Genes, composing a Genome, creating the Agent, and testing it.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xiaoba-dev](https://clawhub.ai/user/xiaoba-dev)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and engineers use this skill to design Rotifer-based agents by decomposing a requested capability into Genes, choosing a composition strategy, creating an Agent, and validating the result.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The skill guides users to install, run, and compose Rotifer Genes that may use filesystem, process, or network access.\n\nMitigation: Review Genes before installing or running them, and avoid sensitive inputs in web-search or external API Genes unless the provider is trusted.\n\nRisk: Agent creation can write local .rotifer agent configuration files and Seq compositions can fail at runtime if adjacent Gene schemas do not match.\n\nMitigation: Inspect generated .rotifer agent configuration files and manually compare adjacent Gene input and output schemas before creating Seq compositions.\n\n## Reference(s):\n\n- [Rotifer Protocol](https://rotifer.dev)\n- [Rotifer Documentation](https://rotifer.dev/docs)\n- [Rotifer Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n- [Rotifer Playground Repository](https://github.com/rotifer-protocol/rotifer-playground)\n\n## Skill Output:\n\n**Output Type(s):** [Guidance, Markdown, Shell commands, Configuration]\n\n**Output Format:** [Markdown with tables and inline bash or JSON code blocks]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May include capability decomposition tables, Gene selection notes, composition commands, Agent validation checklists, and troubleshooting guidance.]\n\n## Skill Version(s):\n\n1.1.0 (source: server release metadata and clawhub.json)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nFile v1.1.0:clawhub.json\n\n{\n \"name\": \"rotifer-agent\",\n \"display_name\": \"Rotifer Agent\",\n \"description\": \"Build AI Agents from Genes — decompose intent into capability units, select the fittest Genes from Arena rankings, compose them with Seq/Par/Cond/Try/TryPool strategies, and validate the pipeline end-to-end.\",\n \"tagline\": \"From intent to running Agent in seven steps\",\n \"version\": \"1.1.0\",\n \"author\": \"rotifer-protocol\",\n \"license\": \"Apache-2.0\",\n \"category\": \"developer-tools\",\n \"tags\": [\n \"rotifer\",\n \"ai-agent\",\n \"agent-builder\",\n \"genome\",\n \"composition\",\n \"pipeline\",\n \"gene\",\n \"orchestration\"\n ],\n \"homepage\": \"https://rotifer.dev\",\n \"repository\": \"https://github.com/rotifer-protocol/rotifer-playground\",\n \"permissions\": [\n \"process:exec\",\n \"filesystem:read\",\n \"filesystem:write\",\n \"network:outbound\"\n ],\n \"screenshots\": [],\n \"demo_video\": null\n}\n\nArchive v1.0.3: 5 files, 7114 bytes\n\nFiles: clawhub.json (794b), README.md (2414b), skill-card.md (2423b), SKILL.md (9583b), _meta.json (132b)\n\nFile v1.0.3:SKILL.md\n\n---\nname: rotifer-agent\ndescription: >-\n End-to-end guide for building AI Agents from Genes: intent decomposition, Gene selection,\n Genome composition, Agent creation, and testing.\n Use when the user mentions \"build Agent\", \"compose Genes\", \"Agent composition\", \"agent create\",\n \"agent run\", \"Genome\", \"assemble Agent\", \"Agent architecture\", \"composition strategy\",\n \"Seq\", \"Par\", \"Cond\", \"Try\", \"TryPool\", \"synthesize Agent\".\n---\n\n# Rotifer Agent — From Genes to Agents\n\nDecompose user intent into capability units, select Genes from the ecosystem, compose a Genome, create and validate an Agent.\n\n## Prerequisites\n\nThis Skill requires the Rotifer CLI:\n\n```bash\nnpx @rotifer/playground --version\n```\n\nOr use the MCP Server for IDE integration:\n\n```json\n{\n \"mcpServers\": {\n \"rotifer\": {\n \"command\": \"npx\",\n \"args\": [\"@rotifer/mcp-server\"]\n }\n }\n}\n```\n\n---\n\n**Hierarchy**: Gene (atomic logic) → Genome (composition) → Agent (runnable entity)\n\n---\n\n## Phase 1: Intent Decomposition\n\nBreak the user's goal into independent capability units (each maps to a Gene).\n\n**Steps**:\n\n1. Confirm the Agent's input and expected output with the user\n2. Decompose the task into 2–6 capability units, each satisfying the Gene three axioms (functional cohesion, self-sufficient interface, independently evaluable)\n3. Label each unit with a domain (e.g. `content.grammar`, `security.audit`)\n4. Confirm the decomposition with the user before proceeding to Phase 2\n\n**Output format**:\n\n| # | Capability unit | Domain | Input | Output |\n|---|----------------|--------|-------|--------|\n| 1 | Grammar check | content.grammar | text | issues[], score |\n| 2 | Readability analysis | content.readability | text | grade, suggestions[] |\n\n---\n\n## Phase 2: Gene Selection\n\nMatch existing Genes to each capability unit.\n\n```bash\nrotifer list\nrotifer arena list --domain \n```\n\n**Selection priority**:\n\n| Priority | Source | Command |\n|----------|--------|---------|\n| 1 | Local Gene with highest Arena rank | `rotifer arena list --domain ` |\n| 2 | Cloud Registry | `rotifer install ` |\n| 3 | Doesn't exist, needs creation | Proceed to Phase 3 |\n\nShow the user candidate Genes' F(g) fitness and fidelity, let them confirm the selection.\n\n---\n\n## Phase 3: Gap Filling\n\nIf a capability unit has no existing Gene:\n\n| Approach | When to use | Action |\n|----------|------------|--------|\n| Create Wrapped Gene | External API / Skill available to wrap | Route to `gene-dev` Skill |\n| Create Native Gene | Pure computation, no external dependencies | Route to `gene-dev` Skill |\n| Adjust decomposition | Capability unit granularity is wrong | Return to Phase 1 |\n| Merge units | Two units are too coupled, splitting makes the interface awkward | Merge into one Gene |\n\nAfter all Genes are ready, proceed to Phase 4.\n\n---\n\n## Phase 4: Genome Composition\n\nChoose a composition strategy based on relationships between capability units.\n\n### Composition Strategy Decision Table\n\n| Strategy | Semantics | Use when | Example |\n|----------|-----------|----------|---------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Previous output feeds the next | Check → Fix → Format |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results | Grammar check + Readability analysis |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input characteristics determine path | Chinese → Chinese proofing / English → English proofing |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable | Main API → Backup API |\n| **TryPool(A, B, C)** | Race: all try, first success wins | Multiple equivalent implementations | Multiple translation services racing |\n\n### Par Merge Strategies\n\nWhen using Par, specify `--par-merge`:\n\n| Strategy | Behavior | Use when |\n|----------|----------|----------|\n| `first` | Take the first completed result | Racing scenario |\n| `concat` | Concatenate all results (array) | Results are complementary |\n| `merge` | Deep-merge objects | Same structure, merge fields |\n\n### Seq Schema Compatibility Warning\n\n> **Known limitation**: Seq composition requires the previous Gene's outputSchema to be compatible with the next Gene's inputSchema. The current version does not auto-validate — schema mismatches cause runtime errors.\n>\n> **Recommendation**: Before creating a Seq composition, manually compare adjacent Genes' inputSchema / outputSchema in phenotype.json to confirm field names and types match.\n\n### Nested Composition\n\nStrategies can be nested:\n\n```text\nSeq(\n Par(grammar-checker, readability-analyzer),\n tone-analyzer\n)\n```\n\nCorresponding CLI:\n\n```bash\nrotifer agent create doc-qa \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Seq\n```\n\n> The current CLI only supports top-level composition strategies. Nested compositions require manual editing of `.rotifer/agents/.json`.\n\n---\n\n## Phase 5: Agent Creation\n\nExecute creation after confirming the composition plan.\n\n### Manual Gene Selection\n\n```bash\nrotifer agent create \\\n --genes \\\n --composition \\\n --par-merge \n```\n\n### Auto-select Genes (by domain ranking)\n\n```bash\nrotifer agent create \\\n --domain \\\n --top \\\n --composition \n```\n\nAfter creation, verify the Agent configuration file `.rotifer/agents/.json` is correct.\n\n---\n\n## Phase 6: Test Run\n\n```bash\nrotifer agent run --input '{\"text\": \"Test input content\"}'\n```\n\n**Validation checklist**:\n\n- Does the output structure match the expected schema?\n- Were all Genes executed? (check logs)\n- Is schema passing correct in Seq composition?\n- Are Par merge results complete?\n- Do error paths (Try/TryPool) degrade correctly?\n\nIf results are unsatisfactory, proceed to Phase 7.\n\n---\n\n## Phase 7: Iterative Optimization\n\n| Problem | Optimization |\n|---------|-------------|\n| One Gene's output quality is poor | `rotifer arena list --domain ` to find alternatives |\n| Seq intermediate results missing fields | Check schema compatibility, consider inserting an adapter Gene |\n| Par merge results are messy | Switch `--par-merge` strategy |\n| Latency too high | Seq → Par (if Genes are independent) |\n| Overall below expectations | Route to `rotifer-arena` Skill for head-to-head Gene evaluation |\n\n---\n\n## Scenario Examples\n\n### Scenario 1: Document Quality Agent\n\n**Goal**: Input text, output grammar issues + readability score + tone analysis.\n\n**Decomposition**:\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Grammar check | grammar-checker | content.grammar |\n| 2 | Readability analysis | readability-analyzer | content.readability |\n| 3 | Tone analysis | tone-analyzer | content.tone |\n\n**Composition**: All three accept `text` input, no dependencies → Par + concat.\n\n```bash\nrotifer agent create doc-quality \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Par \\\n --par-merge concat\n\nrotifer agent run doc-quality --input '{\"text\": \"Document content to check...\"}'\n```\n\n### Scenario 2: Code Review Agent\n\n**Goal**: Input code file, output security vulnerabilities + complexity report + documentation suggestions.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Security audit | security-auditor | security.audit |\n| 2 | Complexity analysis | code-complexity | code.analysis |\n| 3 | Documentation generation | docs-writer | content.docs |\n\n**Composition**: Security audit and complexity analysis can run in parallel, documentation depends on both → Seq(Par(1,2), 3).\n\n```bash\nrotifer agent create code-review \\\n --genes security-auditor code-complexity docs-writer \\\n --composition Seq\n\nrotifer agent run code-review --input '{\"code\": \"...\", \"language\": \"typescript\"}'\n```\n\n> Note: The Par(security-auditor, code-complexity) merged output must be compatible with docs-writer's inputSchema. Manual verification required.\n\n### Scenario 3: Search & Summarize Agent\n\n**Goal**: Input a search query, search → summarize → format output.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Web search | genesis-web-search | search.web |\n| 2 | Text summarization | text-summarizer | content.summarize |\n| 3 | Markdown formatting | markdown-formatter | content.format |\n\n**Composition**: Strict serial pipeline → Seq.\n\n```bash\nrotifer agent create search-digest \\\n --genes genesis-web-search text-summarizer markdown-formatter \\\n --composition Seq\n\nrotifer agent run search-digest --input '{\"query\": \"Rotifer Protocol agent framework\"}'\n```\n\n> Note the Seq schema chain: genesis-web-search output field names must match text-summarizer's inputSchema. Run `cat genes/*/phenotype.json | jq '.inputSchema, .outputSchema'` to verify before creating.\n\n---\n\n## Related Skills\n\n| Skill | Relationship | When to route |\n|-------|-------------|---------------|\n| `gene-dev` | Gene creation/development | Phase 3 gap filling |\n| `rotifer-arena` | Gene comparison & evaluation | Phase 7 when replacing underperforming Genes |\n| `genome` | Genome quality analysis | After Agent creation for overall assessment |\n\n---\n\n## Constraints\n\n- Agent configuration files are stored in `.rotifer/agents/.json` and should not be committed to Git\n- A single Agent should contain 2–6 Genes; more than 6 suggests splitting into multiple Agents\n- Seq schema compatibility is a known limitation — always verify manually before creating\n- Nested compositions require manual JSON editing; the CLI only supports top-level strategies\n\nFile v1.0.3:README.md\n\n# Rotifer Agent\n\n> From intent to running Agent in seven steps.\n\nBuild AI Agents by composing [Genes](https://rotifer.dev/docs) — the atomic, fitness-evaluable capability units of [Rotifer Protocol](https://rotifer.dev).\n\n## What It Does\n\nDescribe what your Agent should do, and this Skill handles the architecture:\n\n| Say this | It does this |\n|----------|-------------|\n| \"Build a document quality checker\" | Decomposes into grammar + readability + tone Genes, recommends Par composition |\n| \"Create a code review agent\" | Selects security-auditor + code-complexity + docs-writer, designs Seq(Par, ...) pipeline |\n| \"Make a search-and-summarize pipeline\" | Picks web-search → summarizer → formatter, wires up Seq |\n\n## The Seven Phases\n\n1. **Intent decomposition** — break the goal into 2-6 capability units\n2. **Gene selection** — find the fittest Gene for each unit from Arena rankings\n3. **Gap filling** — create missing Genes (route to gene-dev)\n4. **Genome composition** — choose Seq / Par / Cond / Try / TryPool\n5. **Agent creation** — `rotifer agent create`\n6. **Test run** — `rotifer agent run` with validation checklist\n7. **Iteration** — swap underperforming Genes, tune composition\n\n## Composition Strategies\n\n| Strategy | Semantics | Use when |\n|----------|-----------|----------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Output of one feeds the next |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input determines the path |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable |\n| **TryPool(A, B, C)** | Race: all try, first wins | Multiple equivalent implementations |\n\n## Quick Start\n\n```\nBuild an agent that checks document quality — grammar, readability, and tone\n```\n\nThe Skill decomposes the intent, finds top-ranked Genes, and creates the Agent.\n\n## Related Skills\n\n- [rotifer-guide](https://clawhub.ai/skills/rotifer-guide) — Learn Rotifer from scratch\n- [rotifer-arena](https://clawhub.ai/skills/rotifer-arena) — Compare and evaluate Genes\n- [rotifer-self-evolving-agent](https://clawhub.ai/skills/rotifer-self-evolving-agent) — Auto-evolve your Agent\n\n## Links\n\n- [Rotifer Protocol](https://rotifer.dev)\n- [Documentation](https://rotifer.dev/docs)\n- [Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n\n## License\n\nApache-2.0\n\nFile v1.0.3:_meta.json\n\n{\n \"ownerId\": \"kn7ec5m4jmk1h4jyzfrxaesjk9838ykv\",\n \"slug\": \"rotifer-agent\",\n \"version\": \"1.0.3\",\n \"publishedAt\": 1776265984118\n}\n\nFile v1.0.3:skill-card.md\n\n## Description:
\nEnd-to-end guide for building AI Agents from Genes: intent decomposition, Gene selection, Genome composition, Agent creation, and testing.
\n\nThis skill is ready for commercial/non-commercial use.
\n\n## Publisher:
\n[xiaoba-dev](https://clawhub.ai/user/xiaoba-dev)
\n\n### License/Terms of Use:
\nMIT-0
\n\n\n## Use Case:
\nDevelopers and engineers use this skill to decompose a desired Rotifer agent into capability units, select Genes, choose a Genome composition strategy, create the agent, and validate its run behavior.
\n\n### Deployment Geography for Use:
\nGlobal
\n\n## Known Risks and Mitigations:
\nRisk: Broad activation wording may cause the skill to be used for general agent-building requests.
\nMitigation: Use explicit Rotifer-specific requests and review the skill guidance before relying on generated plans.
\nRisk: Seq compositions can fail at runtime when adjacent Gene schemas are incompatible.
\nMitigation: Manually compare adjacent Gene inputSchema and outputSchema fields before creating or running a Seq agent.
\n\n\n## Reference(s):
\n- [Rotifer Protocol](https://rotifer.dev)
\n- [Rotifer Documentation](https://rotifer.dev/docs)
\n- [Rotifer Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)
\n- [Rotifer Playground Repository](https://github.com/rotifer-protocol/rotifer-playground)
\n- [Rotifer Guide Skill](https://clawhub.ai/skills/rotifer-guide)
\n- [Rotifer Arena Skill](https://clawhub.ai/skills/rotifer-arena)
\n- [Rotifer Self-Evolving Agent Skill](https://clawhub.ai/skills/rotifer-self-evolving-agent)
\n\n\n## Skill Output:
\n**Output Type(s):** [Guidance, Markdown, Shell commands, Configuration]
\n**Output Format:** [Markdown with tables, checklists, JSON snippets, and inline shell commands]
\n**Output Parameters:** [1D]
\n**Other Properties Related to Output:** [May include Rotifer CLI commands, MCP configuration snippets, and agent composition plans for user review before execution.]
\n\n## Skill Version(s):
\n1.0.3 (source: server release metadata and clawhub.json)
\n\n## Ethical Considerations:
\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.
\n\nFile v1.0.3:clawhub.json\n\n{\n \"name\": \"rotifer-agent\",\n \"display_name\": \"Rotifer Agent\",\n \"description\": \"Build AI Agents from Genes — decompose intent into capability units, select the fittest Genes from Arena rankings, compose them with Seq/Par/Cond/Try/TryPool strategies, and validate the pipeline end-to-end.\",\n \"tagline\": \"From intent to running Agent in seven steps\",\n \"version\": \"1.0.3\",\n \"author\": \"rotifer-protocol\",\n \"license\": \"Apache-2.0\",\n \"category\": \"developer-tools\",\n \"tags\": [\n \"rotifer\",\n \"ai-agent\",\n \"agent-builder\",\n \"genome\",\n \"composition\",\n \"pipeline\",\n \"gene\",\n \"orchestration\"\n ],\n \"homepage\": \"https://rotifer.dev\",\n \"repository\": \"https://github.com/rotifer-protocol/rotifer-playground\",\n \"permissions\": [],\n \"screenshots\": [],\n \"demo_video\": null\n}\n\nArchive v1.0.2: 4 files, 5788 bytes\n\nFiles: clawhub.json (794b), README.md (2414b), SKILL.md (9307b), _meta.json (132b)\n\nFile v1.0.2:SKILL.md\n\n---\nname: rotifer-agent\ndescription: >-\n End-to-end guide for building AI Agents from Genes: intent decomposition, Gene selection,\n Genome composition, Agent creation, and testing.\n Use when the user mentions \"build Agent\", \"compose Genes\", \"Agent composition\", \"agent create\",\n \"agent run\", \"Genome\", \"assemble Agent\", \"Agent architecture\", \"composition strategy\",\n \"Seq\", \"Par\", \"Cond\", \"Try\", \"TryPool\", \"synthesize Agent\".\n---\n\n# Rotifer Agent — From Genes to Agents\n\nDecompose user intent into capability units, select Genes from the ecosystem, compose a Genome, create and validate an Agent.\n\n**Hierarchy**: Gene (atomic logic) → Genome (composition) → Agent (runnable entity)\n\n---\n\n## Phase 1: Intent Decomposition\n\nBreak the user's goal into independent capability units (each maps to a Gene).\n\n**Steps**:\n\n1. Confirm the Agent's input and expected output with the user\n2. Decompose the task into 2–6 capability units, each satisfying the Gene three axioms (functional cohesion, self-sufficient interface, independently evaluable)\n3. Label each unit with a domain (e.g. `content.grammar`, `security.audit`)\n4. Confirm the decomposition with the user before proceeding to Phase 2\n\n**Output format**:\n\n| # | Capability unit | Domain | Input | Output |\n|---|----------------|--------|-------|--------|\n| 1 | Grammar check | content.grammar | text | issues[], score |\n| 2 | Readability analysis | content.readability | text | grade, suggestions[] |\n\n---\n\n## Phase 2: Gene Selection\n\nMatch existing Genes to each capability unit.\n\n```bash\nrotifer list\nrotifer arena list --domain \n```\n\n**Selection priority**:\n\n| Priority | Source | Command |\n|----------|--------|---------|\n| 1 | Local Gene with highest Arena rank | `rotifer arena list --domain ` |\n| 2 | Cloud Registry | `rotifer install ` |\n| 3 | Doesn't exist, needs creation | Proceed to Phase 3 |\n\nShow the user candidate Genes' F(g) fitness and fidelity, let them confirm the selection.\n\n---\n\n## Phase 3: Gap Filling\n\nIf a capability unit has no existing Gene:\n\n| Approach | When to use | Action |\n|----------|------------|--------|\n| Create Wrapped Gene | External API / Skill available to wrap | Route to `gene-dev` Skill |\n| Create Native Gene | Pure computation, no external dependencies | Route to `gene-dev` Skill |\n| Adjust decomposition | Capability unit granularity is wrong | Return to Phase 1 |\n| Merge units | Two units are too coupled, splitting makes the interface awkward | Merge into one Gene |\n\nAfter all Genes are ready, proceed to Phase 4.\n\n---\n\n## Phase 4: Genome Composition\n\nChoose a composition strategy based on relationships between capability units.\n\n### Composition Strategy Decision Table\n\n| Strategy | Semantics | Use when | Example |\n|----------|-----------|----------|---------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Previous output feeds the next | Check → Fix → Format |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results | Grammar check + Readability analysis |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input characteristics determine path | Chinese → Chinese proofing / English → English proofing |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable | Main API → Backup API |\n| **TryPool(A, B, C)** | Race: all try, first success wins | Multiple equivalent implementations | Multiple translation services racing |\n\n### Par Merge Strategies\n\nWhen using Par, specify `--par-merge`:\n\n| Strategy | Behavior | Use when |\n|----------|----------|----------|\n| `first` | Take the first completed result | Racing scenario |\n| `concat` | Concatenate all results (array) | Results are complementary |\n| `merge` | Deep-merge objects | Same structure, merge fields |\n\n### Seq Schema Compatibility Warning\n\n> **Known limitation**: Seq composition requires the previous Gene's outputSchema to be compatible with the next Gene's inputSchema. The current version does not auto-validate — schema mismatches cause runtime errors.\n>\n> **Recommendation**: Before creating a Seq composition, manually compare adjacent Genes' inputSchema / outputSchema in phenotype.json to confirm field names and types match.\n\n### Nested Composition\n\nStrategies can be nested:\n\n```text\nSeq(\n Par(grammar-checker, readability-analyzer),\n tone-analyzer\n)\n```\n\nCorresponding CLI:\n\n```bash\nrotifer agent create doc-qa \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Seq\n```\n\n> The current CLI only supports top-level composition strategies. Nested compositions require manual editing of `.rotifer/agents/.json`.\n\n---\n\n## Phase 5: Agent Creation\n\nExecute creation after confirming the composition plan.\n\n### Manual Gene Selection\n\n```bash\nrotifer agent create \\\n --genes \\\n --composition \\\n --par-merge \n```\n\n### Auto-select Genes (by domain ranking)\n\n```bash\nrotifer agent create \\\n --domain \\\n --top \\\n --composition \n```\n\nAfter creation, verify the Agent configuration file `.rotifer/agents/.json` is correct.\n\n---\n\n## Phase 6: Test Run\n\n```bash\nrotifer agent run --input '{\"text\": \"Test input content\"}'\n```\n\n**Validation checklist**:\n\n- Does the output structure match the expected schema?\n- Were all Genes executed? (check logs)\n- Is schema passing correct in Seq composition?\n- Are Par merge results complete?\n- Do error paths (Try/TryPool) degrade correctly?\n\nIf results are unsatisfactory, proceed to Phase 7.\n\n---\n\n## Phase 7: Iterative Optimization\n\n| Problem | Optimization |\n|---------|-------------|\n| One Gene's output quality is poor | `rotifer arena list --domain ` to find alternatives |\n| Seq intermediate results missing fields | Check schema compatibility, consider inserting an adapter Gene |\n| Par merge results are messy | Switch `--par-merge` strategy |\n| Latency too high | Seq → Par (if Genes are independent) |\n| Overall below expectations | Route to `rotifer-arena` Skill for head-to-head Gene evaluation |\n\n---\n\n## Scenario Examples\n\n### Scenario 1: Document Quality Agent\n\n**Goal**: Input text, output grammar issues + readability score + tone analysis.\n\n**Decomposition**:\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Grammar check | grammar-checker | content.grammar |\n| 2 | Readability analysis | readability-analyzer | content.readability |\n| 3 | Tone analysis | tone-analyzer | content.tone |\n\n**Composition**: All three accept `text` input, no dependencies → Par + concat.\n\n```bash\nrotifer agent create doc-quality \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Par \\\n --par-merge concat\n\nrotifer agent run doc-quality --input '{\"text\": \"Document content to check...\"}'\n```\n\n### Scenario 2: Code Review Agent\n\n**Goal**: Input code file, output security vulnerabilities + complexity report + documentation suggestions.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Security audit | security-auditor | security.audit |\n| 2 | Complexity analysis | code-complexity | code.analysis |\n| 3 | Documentation generation | docs-writer | content.docs |\n\n**Composition**: Security audit and complexity analysis can run in parallel, documentation depends on both → Seq(Par(1,2), 3).\n\n```bash\nrotifer agent create code-review \\\n --genes security-auditor code-complexity docs-writer \\\n --composition Seq\n\nrotifer agent run code-review --input '{\"code\": \"...\", \"language\": \"typescript\"}'\n```\n\n> Note: The Par(security-auditor, code-complexity) merged output must be compatible with docs-writer's inputSchema. Manual verification required.\n\n### Scenario 3: Search & Summarize Agent\n\n**Goal**: Input a search query, search → summarize → format output.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Web search | genesis-web-search | search.web |\n| 2 | Text summarization | text-summarizer | content.summarize |\n| 3 | Markdown formatting | markdown-formatter | content.format |\n\n**Composition**: Strict serial pipeline → Seq.\n\n```bash\nrotifer agent create search-digest \\\n --genes genesis-web-search text-summarizer markdown-formatter \\\n --composition Seq\n\nrotifer agent run search-digest --input '{\"query\": \"Rotifer Protocol agent framework\"}'\n```\n\n> Note the Seq schema chain: genesis-web-search output field names must match text-summarizer's inputSchema. Run `cat genes/*/phenotype.json | jq '.inputSchema, .outputSchema'` to verify before creating.\n\n---\n\n## Related Skills\n\n| Skill | Relationship | When to route |\n|-------|-------------|---------------|\n| `gene-dev` | Gene creation/development | Phase 3 gap filling |\n| `rotifer-arena` | Gene comparison & evaluation | Phase 7 when replacing underperforming Genes |\n| `genome` | Genome quality analysis | After Agent creation for overall assessment |\n\n---\n\n## Constraints\n\n- Agent configuration files are stored in `.rotifer/agents/.json` and should not be committed to Git\n- A single Agent should contain 2–6 Genes; more than 6 suggests splitting into multiple Agents\n- Seq schema compatibility is a known limitation — always verify manually before creating\n- Nested compositions require manual JSON editing; the CLI only supports top-level strategies\n\nFile v1.0.2:README.md\n\n# Rotifer Agent\n\n> From intent to running Agent in seven steps.\n\nBuild AI Agents by composing [Genes](https://rotifer.dev/docs) — the atomic, fitness-evaluable capability units of [Rotifer Protocol](https://rotifer.dev).\n\n## What It Does\n\nDescribe what your Agent should do, and this Skill handles the architecture:\n\n| Say this | It does this |\n|----------|-------------|\n| \"Build a document quality checker\" | Decomposes into grammar + readability + tone Genes, recommends Par composition |\n| \"Create a code review agent\" | Selects security-auditor + code-complexity + docs-writer, designs Seq(Par, ...) pipeline |\n| \"Make a search-and-summarize pipeline\" | Picks web-search → summarizer → formatter, wires up Seq |\n\n## The Seven Phases\n\n1. **Intent decomposition** — break the goal into 2-6 capability units\n2. **Gene selection** — find the fittest Gene for each unit from Arena rankings\n3. **Gap filling** — create missing Genes (route to gene-dev)\n4. **Genome composition** — choose Seq / Par / Cond / Try / TryPool\n5. **Agent creation** — `rotifer agent create`\n6. **Test run** — `rotifer agent run` with validation checklist\n7. **Iteration** — swap underperforming Genes, tune composition\n\n## Composition Strategies\n\n| Strategy | Semantics | Use when |\n|----------|-----------|----------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Output of one feeds the next |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input determines the path |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable |\n| **TryPool(A, B, C)** | Race: all try, first wins | Multiple equivalent implementations |\n\n## Quick Start\n\n```\nBuild an agent that checks document quality — grammar, readability, and tone\n```\n\nThe Skill decomposes the intent, finds top-ranked Genes, and creates the Agent.\n\n## Related Skills\n\n- [rotifer-guide](https://clawhub.ai/skills/rotifer-guide) — Learn Rotifer from scratch\n- [rotifer-arena](https://clawhub.ai/skills/rotifer-arena) — Compare and evaluate Genes\n- [rotifer-self-evolving-agent](https://clawhub.ai/skills/rotifer-self-evolving-agent) — Auto-evolve your Agent\n\n## Links\n\n- [Rotifer Protocol](https://rotifer.dev)\n- [Documentation](https://rotifer.dev/docs)\n- [Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n\n## License\n\nApache-2.0\n\nFile v1.0.2:_meta.json\n\n{\n \"ownerId\": \"kn7ec5m4jmk1h4jyzfrxaesjk9838ykv\",\n \"slug\": \"rotifer-agent\",\n \"version\": \"1.0.2\",\n \"publishedAt\": 1776261766520\n}\n\nFile v1.0.2:clawhub.json\n\n{\n \"name\": \"rotifer-agent\",\n \"display_name\": \"Rotifer Agent\",\n \"description\": \"Build AI Agents from Genes — decompose intent into capability units, select the fittest Genes from Arena rankings, compose them with Seq/Par/Cond/Try/TryPool strategies, and validate the pipeline end-to-end.\",\n \"tagline\": \"From intent to running Agent in seven steps\",\n \"version\": \"1.0.2\",\n \"author\": \"rotifer-protocol\",\n \"license\": \"Apache-2.0\",\n \"category\": \"developer-tools\",\n \"tags\": [\n \"rotifer\",\n \"ai-agent\",\n \"agent-builder\",\n \"genome\",\n \"composition\",\n \"pipeline\",\n \"gene\",\n \"orchestration\"\n ],\n \"homepage\": \"https://rotifer.dev\",\n \"repository\": \"https://github.com/rotifer-protocol/rotifer-playground\",\n \"permissions\": [],\n \"screenshots\": [],\n \"demo_video\": null\n}\n\nArchive v1.0.1: 4 files, 5787 bytes\n\nFiles: clawhub.json (794b), README.md (2414b), SKILL.md (9303b), _meta.json (132b)\n\nFile v1.0.1:SKILL.md\n\n---\nname: rotifer-agent\ndescription: >-\n End-to-end guide for building AI Agents from Genes: intent decomposition, Gene selection,\n Genome composition, Agent creation, and testing.\n Use when the user mentions \"build Agent\", \"compose Genes\", \"Agent composition\", \"agent create\",\n \"agent run\", \"Genome\", \"assemble Agent\", \"Agent architecture\", \"composition strategy\",\n \"Seq\", \"Par\", \"Cond\", \"Try\", \"TryPool\", \"synthesize Agent\".\n---\n\n# Rotifer Agent — From Genes to Agents\n\nDecompose user intent into capability units, select Genes from the ecosystem, compose a Genome, create and validate an Agent.\n\n**Hierarchy**: Gene (atomic logic) → Genome (composition) → Agent (runnable entity)\n\n---\n\n## Phase 1: Intent Decomposition\n\nBreak the user's goal into independent capability units (each maps to a Gene).\n\n**Steps**:\n\n1. Confirm the Agent's input and expected output with the user\n2. Decompose the task into 2–6 capability units, each satisfying the Gene three axioms (functional cohesion, self-sufficient interface, independently evaluable)\n3. Label each unit with a domain (e.g. `content.grammar`, `security.audit`)\n4. Confirm the decomposition with the user before proceeding to Phase 2\n\n**Output format**:\n\n| # | Capability unit | Domain | Input | Output |\n|---|----------------|--------|-------|--------|\n| 1 | Grammar check | content.grammar | text | issues[], score |\n| 2 | Readability analysis | content.readability | text | grade, suggestions[] |\n\n---\n\n## Phase 2: Gene Selection\n\nMatch existing Genes to each capability unit.\n\n```bash\nrotifer list\nrotifer arena list --domain \n```\n\n**Selection priority**:\n\n| Priority | Source | Command |\n|----------|--------|---------|\n| 1 | Local Gene with highest Arena rank | `rotifer arena list --domain ` |\n| 2 | Cloud Registry | `rotifer install ` |\n| 3 | Doesn't exist, needs creation | Proceed to Phase 3 |\n\nShow the user candidate Genes' F(g) fitness and fidelity, let them confirm the selection.\n\n---\n\n## Phase 3: Gap Filling\n\nIf a capability unit has no existing Gene:\n\n| Approach | When to use | Action |\n|----------|------------|--------|\n| Create Wrapped Gene | External API / Skill available to wrap | Route to `gene-dev` Skill |\n| Create Native Gene | Pure computation, no external dependencies | Route to `gene-dev` Skill |\n| Adjust decomposition | Capability unit granularity is wrong | Return to Phase 1 |\n| Merge units | Two units are too coupled, splitting makes the interface awkward | Merge into one Gene |\n\nAfter all Genes are ready, proceed to Phase 4.\n\n---\n\n## Phase 4: Genome Composition\n\nChoose a composition strategy based on relationships between capability units.\n\n### Composition Strategy Decision Table\n\n| Strategy | Semantics | Use when | Example |\n|----------|-----------|----------|---------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Previous output feeds the next | Check → Fix → Format |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results | Grammar check + Readability analysis |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input characteristics determine path | Chinese → Chinese proofing / English → English proofing |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable | Main API → Backup API |\n| **TryPool(A, B, C)** | Race: all try, first success wins | Multiple equivalent implementations | Multiple translation services racing |\n\n### Par Merge Strategies\n\nWhen using Par, specify `--par-merge`:\n\n| Strategy | Behavior | Use when |\n|----------|----------|----------|\n| `first` | Take the first completed result | Racing scenario |\n| `concat` | Concatenate all results (array) | Results are complementary |\n| `merge` | Deep-merge objects | Same structure, merge fields |\n\n### Seq Schema Compatibility Warning\n\n> **Known limitation**: Seq composition requires the previous Gene's outputSchema to be compatible with the next Gene's inputSchema. The current version does not auto-validate — schema mismatches cause runtime errors.\n>\n> **Recommendation**: Before creating a Seq composition, manually compare adjacent Genes' inputSchema / outputSchema in phenotype.json to confirm field names and types match.\n\n### Nested Composition\n\nStrategies can be nested:\n\n```\nSeq(\n Par(grammar-checker, readability-analyzer),\n tone-analyzer\n)\n```\n\nCorresponding CLI:\n\n```bash\nrotifer agent create doc-qa \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Seq\n```\n\n> The current CLI only supports top-level composition strategies. Nested compositions require manual editing of `.rotifer/agents/.json`.\n\n---\n\n## Phase 5: Agent Creation\n\nExecute creation after confirming the composition plan.\n\n### Manual Gene Selection\n\n```bash\nrotifer agent create \\\n --genes \\\n --composition \\\n --par-merge \n```\n\n### Auto-select Genes (by domain ranking)\n\n```bash\nrotifer agent create \\\n --domain \\\n --top \\\n --composition \n```\n\nAfter creation, verify the Agent configuration file `.rotifer/agents/.json` is correct.\n\n---\n\n## Phase 6: Test Run\n\n```bash\nrotifer agent run --input '{\"text\": \"Test input content\"}'\n```\n\n**Validation checklist**:\n\n- Does the output structure match the expected schema?\n- Were all Genes executed? (check logs)\n- Is schema passing correct in Seq composition?\n- Are Par merge results complete?\n- Do error paths (Try/TryPool) degrade correctly?\n\nIf results are unsatisfactory, proceed to Phase 7.\n\n---\n\n## Phase 7: Iterative Optimization\n\n| Problem | Optimization |\n|---------|-------------|\n| One Gene's output quality is poor | `rotifer arena list --domain ` to find alternatives |\n| Seq intermediate results missing fields | Check schema compatibility, consider inserting an adapter Gene |\n| Par merge results are messy | Switch `--par-merge` strategy |\n| Latency too high | Seq → Par (if Genes are independent) |\n| Overall below expectations | Route to `rotifer-arena` Skill for head-to-head Gene evaluation |\n\n---\n\n## Scenario Examples\n\n### Scenario 1: Document Quality Agent\n\n**Goal**: Input text, output grammar issues + readability score + tone analysis.\n\n**Decomposition**:\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Grammar check | grammar-checker | content.grammar |\n| 2 | Readability analysis | readability-analyzer | content.readability |\n| 3 | Tone analysis | tone-analyzer | content.tone |\n\n**Composition**: All three accept `text` input, no dependencies → Par + concat.\n\n```bash\nrotifer agent create doc-quality \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Par \\\n --par-merge concat\n\nrotifer agent run doc-quality --input '{\"text\": \"Document content to check...\"}'\n```\n\n### Scenario 2: Code Review Agent\n\n**Goal**: Input code file, output security vulnerabilities + complexity report + documentation suggestions.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Security audit | security-auditor | security.audit |\n| 2 | Complexity analysis | code-complexity | code.analysis |\n| 3 | Documentation generation | docs-writer | content.docs |\n\n**Composition**: Security audit and complexity analysis can run in parallel, documentation depends on both → Seq(Par(1,2), 3).\n\n```bash\nrotifer agent create code-review \\\n --genes security-auditor code-complexity docs-writer \\\n --composition Seq\n\nrotifer agent run code-review --input '{\"code\": \"...\", \"language\": \"typescript\"}'\n```\n\n> Note: The Par(security-auditor, code-complexity) merged output must be compatible with docs-writer's inputSchema. Manual verification required.\n\n### Scenario 3: Search & Summarize Agent\n\n**Goal**: Input a search query, search → summarize → format output.\n\n| # | Capability | Gene | Domain |\n|---|-----------|------|--------|\n| 1 | Web search | genesis-web-search | search.web |\n| 2 | Text summarization | text-summarizer | content.summarize |\n| 3 | Markdown formatting | markdown-formatter | content.format |\n\n**Composition**: Strict serial pipeline → Seq.\n\n```bash\nrotifer agent create search-digest \\\n --genes genesis-web-search text-summarizer markdown-formatter \\\n --composition Seq\n\nrotifer agent run search-digest --input '{\"query\": \"Rotifer Protocol agent framework\"}'\n```\n\n> Note the Seq schema chain: genesis-web-search output field names must match text-summarizer's inputSchema. Run `cat genes/*/phenotype.json | jq '.inputSchema, .outputSchema'` to verify before creating.\n\n---\n\n## Related Skills\n\n| Skill | Relationship | When to route |\n|-------|-------------|---------------|\n| `gene-dev` | Gene creation/development | Phase 3 gap filling |\n| `rotifer-arena` | Gene comparison & evaluation | Phase 7 when replacing underperforming Genes |\n| `genome` | Genome quality analysis | After Agent creation for overall assessment |\n\n---\n\n## Constraints\n\n- Agent configuration files are stored in `.rotifer/agents/.json` and should not be committed to Git\n- A single Agent should contain 2–6 Genes; more than 6 suggests splitting into multiple Agents\n- Seq schema compatibility is a known limitation — always verify manually before creating\n- Nested compositions require manual JSON editing; the CLI only supports top-level strategies\n\nFile v1.0.1:README.md\n\n# Rotifer Agent\n\n> From intent to running Agent in seven steps.\n\nBuild AI Agents by composing [Genes](https://rotifer.dev/docs) — the atomic, fitness-evaluable capability units of [Rotifer Protocol](https://rotifer.dev).\n\n## What It Does\n\nDescribe what your Agent should do, and this Skill handles the architecture:\n\n| Say this | It does this |\n|----------|-------------|\n| \"Build a document quality checker\" | Decomposes into grammar + readability + tone Genes, recommends Par composition |\n| \"Create a code review agent\" | Selects security-auditor + code-complexity + docs-writer, designs Seq(Par, ...) pipeline |\n| \"Make a search-and-summarize pipeline\" | Picks web-search → summarizer → formatter, wires up Seq |\n\n## The Seven Phases\n\n1. **Intent decomposition** — break the goal into 2-6 capability units\n2. **Gene selection** — find the fittest Gene for each unit from Arena rankings\n3. **Gap filling** — create missing Genes (route to gene-dev)\n4. **Genome composition** — choose Seq / Par / Cond / Try / TryPool\n5. **Agent creation** — `rotifer agent create`\n6. **Test run** — `rotifer agent run` with validation checklist\n7. **Iteration** — swap underperforming Genes, tune composition\n\n## Composition Strategies\n\n| Strategy | Semantics | Use when |\n|----------|-----------|----------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Output of one feeds the next |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input determines the path |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable |\n| **TryPool(A, B, C)** | Race: all try, first wins | Multiple equivalent implementations |\n\n## Quick Start\n\n```\nBuild an agent that checks document quality — grammar, readability, and tone\n```\n\nThe Skill decomposes the intent, finds top-ranked Genes, and creates the Agent.\n\n## Related Skills\n\n- [rotifer-guide](https://clawhub.ai/skills/rotifer-guide) — Learn Rotifer from scratch\n- [rotifer-arena](https://clawhub.ai/skills/rotifer-arena) — Compare and evaluate Genes\n- [rotifer-self-evolving-agent](https://clawhub.ai/skills/rotifer-self-evolving-agent) — Auto-evolve your Agent\n\n## Links\n\n- [Rotifer Protocol](https://rotifer.dev)\n- [Documentation](https://rotifer.dev/docs)\n- [Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n\n## License\n\nApache-2.0\n\nFile v1.0.1:_meta.json\n\n{\n \"ownerId\": \"kn7ec5m4jmk1h4jyzfrxaesjk9838ykv\",\n \"slug\": \"rotifer-agent\",\n \"version\": \"1.0.1\",\n \"publishedAt\": 1776261574356\n}\n\nFile v1.0.1:clawhub.json\n\n{\n \"name\": \"rotifer-agent\",\n \"display_name\": \"Rotifer Agent\",\n \"description\": \"Build AI Agents from Genes — decompose intent into capability units, select the fittest Genes from Arena rankings, compose them with Seq/Par/Cond/Try/TryPool strategies, and validate the pipeline end-to-end.\",\n \"tagline\": \"From intent to running Agent in seven steps\",\n \"version\": \"1.0.1\",\n \"author\": \"rotifer-protocol\",\n \"license\": \"Apache-2.0\",\n \"category\": \"developer-tools\",\n \"tags\": [\n \"rotifer\",\n \"ai-agent\",\n \"agent-builder\",\n \"genome\",\n \"composition\",\n \"pipeline\",\n \"gene\",\n \"orchestration\"\n ],\n \"homepage\": \"https://rotifer.dev\",\n \"repository\": \"https://github.com/rotifer-protocol/rotifer-playground\",\n \"permissions\": [],\n \"screenshots\": [],\n \"demo_video\": null\n}\n\nArchive v1.0.0: 4 files, 5857 bytes\n\nFiles: clawhub.json (794b), README.md (2414b), SKILL.md (8626b), _meta.json (132b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: rotifer-agent\ndescription: >-\n 从 Gene 构建 Agent 的全流程架构指南：意图分解、Gene 选型、Genome 编排、Agent 创建与测试。\n 当用户提到 \"构建 Agent\" \"组合 Gene\" \"Agent 编排\" \"agent create\" \"agent run\"\n \"Genome 编排\" \"组装 Agent\" \"Agent 架构\" \"编排策略\" \"Seq\" \"Par\" \"Cond\" \"Try\"\n \"TryPool\" \"合成 Agent\" 时使用。\n---\n\n# Rotifer Agent — 从 Gene 到 Agent\n\n将用户意图分解为能力单元，从 Gene 生态中选型、编排 Genome、创建并验证 Agent。\n\n**层级关系**：Gene（原子逻辑） -> Genome（组合编排） -> Agent（可运行实体）\n\n---\n\n## Phase 1：意图分解\n\n将用户描述的目标拆解为独立的能力单元（每个单元对应一个 Gene）。\n\n**操作步骤**：\n\n1. 向用户确认 Agent 的输入和期望输出\n2. 将任务拆解为 2-6 个能力单元，每个能力单元满足 Gene 三公理（功能内聚、接口自足、独立可评估）\n3. 为每个能力单元标注 domain（如 `content.grammar`、`security.audit`）\n4. 向用户确认拆解方案后进入 Phase 2\n\n**输出格式**：\n\n| # | 能力单元 | domain | 输入 | 输出 |\n|---|---------|--------|------|------|\n| 1 | 语法检查 | content.grammar | text | issues[], score |\n| 2 | 可读性分析 | content.readability | text | grade, suggestions[] |\n\n---\n\n## Phase 2：Gene 选型\n\n为每个能力单元匹配现有 Gene。\n\n```bash\nrotifer list\nrotifer arena list --domain \n```\n\n**选型优先级**：\n\n| 优先级 | 来源 | 命令 |\n|--------|------|------|\n| 1 | 本地已有 Gene（Arena 排名最高） | `rotifer arena list --domain ` |\n| 2 | Cloud Registry | `rotifer install ` |\n| 3 | 不存在，需要创建 | 进入 Phase 3 |\n\n向用户展示候选 Gene 的 F(g) 适应度和 fidelity，由用户确认选择。\n\n---\n\n## Phase 3：缺口补足\n\n如果某个能力单元没有现成 Gene：\n\n| 方案 | 适用场景 | 操作 |\n|------|---------|------|\n| 创建 Wrapped Gene | 已有外部 API / Skill 可包装 | 路由到 `gene-dev` Skill |\n| 创建 Native Gene | 纯计算逻辑，无外部依赖 | 路由到 `gene-dev` Skill |\n| 调整拆解方案 | 能力单元粒度不当 | 返回 Phase 1 重新拆解 |\n| 合并能力单元 | 两个单元耦合度高，拆分后接口笨拙 | 合并为一个 Gene |\n\n完成所有 Gene 准备后进入 Phase 4。\n\n---\n\n## Phase 4：Genome 编排\n\n根据能力单元之间的关系选择组合策略。\n\n### 编排策略决策表\n\n| 策略 | 语义 | 适用场景 | 示例 |\n|------|------|---------|------|\n| **Seq(A, B, C)** | 串行管道：A -> B -> C | 前序输出是后序输入 | 检查 -> 修复 -> 格式化 |\n| **Par(A, B)** | 并行聚合：A 和 B 同时执行 | 各自独立，结果需合并 | 语法检查 + 可读性分析 |\n| **Cond(p, A, B)** | 条件分支：p 为真走 A，否则走 B | 输入特征决定处理路径 | 中文 -> 中文校对 / 英文 -> 英文校对 |\n| **Try(A, B)** | 降级兜底：A 失败则 B | 主路径不可靠，需备选 | 主 API -> 备用 API |\n| **TryPool(A, B, C)** | 竞速池：全部尝试，取首个成功 | 多个等价实现，优先最快 | 多个翻译服务竞速 |\n\n### Par 合并策略\n\n使用 Par 时需指定 `--par-merge`：\n\n| 策略 | 行为 | 适用场景 |\n|------|------|---------|\n| `first` | 取第一个完成的结果 | 竞速场景 |\n| `concat` | 拼接所有结果（数组） | 结果互补，需全部保留 |\n| `merge` | 深度合并对象 | 结果结构相同，需合并字段 |\n\n### Seq Schema 兼容性警告\n\n> **已知限制**：Seq 编排要求前序 Gene 的 outputSchema 与后序 Gene 的 inputSchema 兼容。当前版本不做自动校验，schema 不匹配会导致运行时错误。\n>\n> **建议**：创建 Seq 编排前，手动比对相邻 Gene 的 phenotype.json 中的 inputSchema / outputSchema，确认字段名称和类型一致。\n\n### 嵌套编排\n\n策略可嵌套组合：\n\n```\nSeq(\n Par(grammar-checker, readability-analyzer),\n tone-analyzer\n)\n```\n\n对应 CLI：\n\n```bash\nrotifer agent create doc-qa \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Seq\n```\n\n> 当前 CLI 仅支持顶层编排策略。嵌套编排需手动编辑 `.rotifer/agents/.json`。\n\n---\n\n## Phase 5：Agent 创建\n\n确认编排方案后执行创建。\n\n### 手动选 Gene\n\n```bash\nrotifer agent create \\\n --genes \\\n --composition \\\n --par-merge \n```\n\n### 自动选 Gene（按 domain 排名）\n\n```bash\nrotifer agent create \\\n --domain \\\n --top \\\n --composition \n```\n\n创建后确认 Agent 配置文件 `.rotifer/agents/.json` 内容正确。\n\n---\n\n## Phase 6：测试运行\n\n```bash\nrotifer agent run --input '{\"text\": \"测试输入内容\"}'\n```\n\n**验证清单**：\n\n- 输出结构是否符合预期 schema\n- 各 Gene 是否均被执行（检查日志）\n- Seq 编排中 schema 传递是否正确\n- Par 编排中合并结果是否完整\n- 错误路径（Try/TryPool）是否正确降级\n\n如果结果不理想，进入 Phase 7。\n\n---\n\n## Phase 7：迭代优化\n\n| 问题 | 优化方向 |\n|------|---------|\n| 某个 Gene 输出质量差 | `rotifer arena list --domain ` 查找替代 Gene |\n| Seq 中间结果丢失字段 | 检查 schema 兼容性，考虑插入适配 Gene |\n| Par 合并结果混乱 | 切换 `--par-merge` 策略 |\n| 延迟过高 | Seq -> Par（如果 Gene 间无依赖） |\n| 整体不及预期 | 路由到 `rotifer-arena` Skill 对比评估各 Gene |\n\n---\n\n## 示例场景\n\n### 场景 1：文档质量检查 Agent\n\n**目标**：输入文本，输出语法问题 + 可读性评分 + 语气分析。\n\n**意图分解**：\n\n| # | 能力单元 | Gene | domain |\n|---|---------|------|--------|\n| 1 | 语法检查 | grammar-checker | content.grammar |\n| 2 | 可读性分析 | readability-analyzer | content.readability |\n| 3 | 语气分析 | tone-analyzer | content.tone |\n\n**编排**：三个 Gene 均接受 `text` 输入，互不依赖 -> Par + concat。\n\n```bash\nrotifer agent create doc-quality \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Par \\\n --par-merge concat\n\nrotifer agent run doc-quality --input '{\"text\": \"待检查的文档内容...\"}'\n```\n\n### 场景 2：代码审查 Agent\n\n**目标**：输入代码文件，输出安全漏洞 + 复杂度报告 + 文档建议。\n\n| # | 能力单元 | Gene | domain |\n|---|---------|------|--------|\n| 1 | 安全审计 | security-auditor | security.audit |\n| 2 | 复杂度分析 | code-complexity | code.analysis |\n| 3 | 文档生成 | docs-writer | content.docs |\n\n**编排**：安全审计和复杂度分析可并行，文档生成依赖前两者结果 -> Seq(Par(1,2), 3)。\n\n```bash\nrotifer agent create code-review \\\n --genes security-auditor code-complexity docs-writer \\\n --composition Seq\n\nrotifer agent run code-review --input '{\"code\": \"...\", \"language\": \"typescript\"}'\n```\n\n> 注意：此 Seq 编排中 Par(security-auditor, code-complexity) 的合并输出需与 docs-writer 的 inputSchema 兼容。当前需手动验证。\n\n### 场景 3：搜索摘要 Agent\n\n**目标**：输入查询关键词，搜索 -> 摘要 -> 格式化输出。\n\n| # | 能力单元 | Gene | domain |\n|---|---------|------|--------|\n| 1 | Web 搜索 | genesis-web-search | search.web |\n| 2 | 文本摘要 | text-summarizer | content.summarize |\n| 3 | Markdown 格式化 | markdown-formatter | content.format |\n\n**编排**：严格串行管道 -> Seq。\n\n```bash\nrotifer agent create search-digest \\\n --genes genesis-web-search text-summarizer markdown-formatter \\\n --composition Seq\n\nrotifer agent run search-digest --input '{\"query\": \"Rotifer Protocol agent framework\"}'\n```\n\n> 注意 Seq schema 链：genesis-web-search 输出的字段名必须匹配 text-summarizer 的 inputSchema。建议创建前执行 `cat genes/*/phenotype.json | jq '.inputSchema, .outputSchema'` 比对。\n\n---\n\n## 关联 Skill\n\n| Skill | 关系 | 路由时机 |\n|-------|------|---------|\n| `gene-dev` | Gene 创建/开发 | Phase 3 缺口补足时 |\n| `rotifer-arena` | Gene 对比评估 | Phase 7 需要替换 Gene 时 |\n| `genome` | Genome 评估分析 | Agent 创建后评估整体质量 |\n\n---\n\n## 约束\n\n- Agent 配置文件存储于 `.rotifer/agents/.json`，不纳入 Git 版本控制\n- 单个 Agent 包含的 Gene 数量建议 2-6 个；超过 6 个应考虑拆分为多个 Agent\n- Seq 编排的 schema 兼容性是已知限制，创建前务必手动验证\n- 嵌套编排需手动编辑 JSON 配置，CLI 仅支持顶层策略\n\nFile v1.0.0:README.md\n\n# Rotifer Agent\n\n> From intent to running Agent in seven steps.\n\nBuild AI Agents by composing [Genes](https://rotifer.dev/docs) — the atomic, fitness-evaluable capability units of [Rotifer Protocol](https://rotifer.dev).\n\n## What It Does\n\nDescribe what your Agent should do, and this Skill handles the architecture:\n\n| Say this | It does this |\n|----------|-------------|\n| \"Build a document quality checker\" | Decomposes into grammar + readability + tone Genes, recommends Par composition |\n| \"Create a code review agent\" | Selects security-auditor + code-complexity + docs-writer, designs Seq(Par, ...) pipeline |\n| \"Make a search-and-summarize pipeline\" | Picks web-search → summarizer → formatter, wires up Seq |\n\n## The Seven Phases\n\n1. **Intent decomposition** — break the goal into 2-6 capability units\n2. **Gene selection** — find the fittest Gene for each unit from Arena rankings\n3. **Gap filling** — create missing Genes (route to gene-dev)\n4. **Genome composition** — choose Seq / Par / Cond / Try / TryPool\n5. **Agent creation** — `rotifer agent create`\n6. **Test run** — `rotifer agent run` with validation checklist\n7. **Iteration** — swap underperforming Genes, tune composition\n\n## Composition Strategies\n\n| Strategy | Semantics | Use when |\n|----------|-----------|----------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Output of one feeds the next |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input determines the path |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable |\n| **TryPool(A, B, C)** | Race: all try, first wins | Multiple equivalent implementations |\n\n## Quick Start\n\n```\nBuild an agent that checks document quality — grammar, readability, and tone\n```\n\nThe Skill decomposes the intent, finds top-ranked Genes, and creates the Agent.\n\n## Related Skills\n\n- [rotifer-guide](https://clawhub.ai/skills/rotifer-guide) — Learn Rotifer from scratch\n- [rotifer-arena](https://clawhub.ai/skills/rotifer-arena) — Compare and evaluate Genes\n- [rotifer-self-evolving-agent](https://clawhub.ai/skills/rotifer-self-evolving-agent) — Auto-evolve your Agent\n\n## Links\n\n- [Rotifer Protocol](https://rotifer.dev)\n- [Documentation](https://rotifer.dev/docs)\n- [Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n\n## License\n\nApache-2.0\n\nFile v1.0.0:_meta.json\n\n{\n \"ownerId\": \"kn7ec5m4jmk1h4jyzfrxaesjk9838ykv\",\n \"slug\": \"rotifer-agent\",\n \"version\": \"1.0.0\",\n \"publishedAt\": 1776261194798\n}\n\nFile v1.0.0:clawhub.json\n\n{\n \"name\": \"rotifer-agent\",\n \"display_name\": \"Rotifer Agent\",\n \"description\": \"Build AI Agents from Genes — decompose intent into capability units, select the fittest Genes from Arena rankings, compose them with Seq/Par/Cond/Try/TryPool strategies, and validate the pipeline end-to-end.\",\n \"tagline\": \"From intent to running Agent in seven steps\",\n \"version\": \"1.0.0\",\n \"author\": \"rotifer-protocol\",\n \"license\": \"Apache-2.0\",\n \"category\": \"developer-tools\",\n \"tags\": [\n \"rotifer\",\n \"ai-agent\",\n \"agent-builder\",\n \"genome\",\n \"composition\",\n \"pipeline\",\n \"gene\",\n \"orchestration\"\n ],\n \"homepage\": \"https://rotifer.dev\",\n \"repository\": \"https://github.com/rotifer-protocol/rotifer-playground\",\n \"permissions\": [],\n \"screenshots\": [],\n \"demo_video\": null\n}","readmeExcerpt":"Skill: rotifer-agent Owner: xiaoba-dev Summary: Build a Rotifer Agent from Genes — decompose intent into capability units, pick Genes by Arena F(g), compose a Rotifer Genome, create the Agent and test it. Use when the user is working with Rotifer specifically: \"Rotifer Agent\", \"compose Genes\", \"Rotifer Genome\", \"Gene composition\", \"rotifer agent create\". Do NOT use for general agent frameworks, or when the words Seq ","codeSnippets":[],"executableExamples":[{"language":"bash","snippet":"npx @rotifer/playground --version\nrotifer doctor"},{"language":"json","snippet":"{\n \"mcpServers\": {\n \"rotifer\": {\n \"command\": \"npx\",\n \"args\": [\"@rotifer/mcp-server\"]\n }\n }\n}"},{"language":"bash","snippet":"rotifer list # what is already in this workspace\nrotifer arena list --domain # ranked by F(g) within one domain\nrotifer search --domain # the Cloud registry — everyone else's Genes"},{"language":"bash","snippet":"rotifer run --input '{\"...\": \"...\"}'"},{"language":"text","snippet":"Seq(\n Par(grammar-checker, readability-analyzer),\n tone-analyzer\n)"},{"language":"bash","snippet":"rotifer agent create doc-qa \\\n --genes grammar-checker readability-analyzer tone-analyzer \\\n --composition Seq"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: rotifer-agent\ndescription: >-\n Build a Rotifer Agent from Genes — decompose intent into capability units, pick Genes by\n Arena F(g), compose a Rotifer Genome, create the Agent and test it.\n Use when the user is working with Rotifer specifically: \"Rotifer Agent\", \"compose Genes\",\n \"Rotifer Genome\", \"Gene composition\", \"rotifer agent create\".\n Do NOT use for general agent frameworks, or when the words Seq / Par / Cond / Try appear as\n ordinary programming terms — in Rotifer those name Genome composition strategies, and this\n Skill applies only when the user is composing Rotifer Genes.\n---\n\n# Rotifer Agent — From Genes to Agents\n\nDecompose user intent into capability units, select Genes from the ecosystem, compose a Genome, create and validate an Agent.\n\n## Prerequisites\n\nThis Skill requires the Rotifer CLI:\n\n```bash\nnpx @rotifer/playground --version\nrotifer doctor\n```\n\n`rotifer doctor` checks the TypeScript→WASM toolchain. Composing an Agent from\nNative Genes means compiling them, and without esbuild and javy that step fails\nwith an error that reads like a code problem.\n\nOr use the MCP Server for IDE integration:\n\n```json\n{\n \"mcpServers\": {\n \"rotifer\": {\n \"command\": \"npx\",\n \"args\": [\"@rotifer/mcp-server\"]\n }\n }\n}\n```\n\n---\n\n**Hierarchy**: Gene (atomic logic) → Genome (composition) → Agent (runnable entity)\n\n---\n\n## Phase 1: Intent Decomposition\n\nBreak the user's goal into independent capability units (each maps to a Gene).\n\n**Steps**:\n\n1. Confirm the Agent's input and expected output with the user\n2. Decompose the task into 2–6 capability units, each satisfying the Gene three axioms (functional cohesion, self-sufficient interface, independently evaluable)\n3. Label each unit with a domain (e.g. `content.grammar`, `security.audit`)\n4. Confirm the decomposition with the user before proceeding to Phase 2\n\n**Output format**:\n\n| # | Capability unit | Domain | Input | Output |\n|---|----------------|--------|-------|--------|\n| 1 | Grammar check | content.grammar | text | issues[], score |\n| 2 | Readability analysis | content.readability | text | grade, suggestions[] |\n\n---\n\n## Phase 2: Gene Selection\n\nMatch existing Genes to each capability unit.\n\n```bash\nrotifer list # what is already in this workspace\nrotifer arena list --domain # ranked by F(g) within one domain\nrotifer search --domain # the Cloud registry — everyone else's Genes\n```\n\nThe three answer different questions, and skipping the third is how a Genome\nends up built only from what happened to be installed already.\n\n**Selection priority**:\n\n| Priority | Source | Command |\n|----------|--------|---------|\n| 1 | Local Gene with highest Arena rank | `rotifer arena list --domain ` |\n| 2 | Cloud Registry | `rotifer search ` → `rotifer info ` → `rotifer install ` |\n| 3 | Doesn't exist, needs creation | Proceed to Phase 3 |\n\nBefore committing a candidate to the Genome, run "},{"path":"README.md","content":"# Rotifer Agent\n\n> From intent to running Agent in seven steps.\n\nBuild a Rotifer Agent by composing [Genes](https://rotifer.dev/docs) — the atomic, fitness-evaluable capability units of [Rotifer Protocol](https://rotifer.dev).\n\n**Not a general agent framework.** This composes Rotifer Genes into a Rotifer\nGenome, through the Rotifer CLI. It does not build agents for LangChain, CrewAI\nor anything else, and the words Seq / Par / Cond / Try name Genome composition\nstrategies here — seeing them in ordinary code is not a reason to invoke it.\n\n## What It Does\n\nSay what the Agent should do **in Rotifer terms**, and this Skill handles the architecture:\n\n| Say this | It does this |\n|----------|-------------|\n| \"Build a Rotifer Agent that checks document quality\" | Decomposes into grammar + readability + tone Genes, recommends Par composition |\n| \"Compose Rotifer Genes into a code review Agent\" | Selects security-auditor + code-complexity + docs-writer, designs a Seq(Par, …) Genome |\n| \"Make a Rotifer Genome for search and summarize\" | Picks web-search → summarizer → formatter, wires up Seq |\n\n## The Seven Phases\n\n1. **Intent decomposition** — break the goal into 2-6 capability units\n2. **Gene selection** — find the fittest Gene for each unit from Arena rankings\n3. **Gap filling** — create missing Genes (route to gene → dev module)\n4. **Genome composition** — choose Seq / Par / Cond / Try / TryPool\n5. **Agent creation** — `rotifer agent create`\n6. **Test run** — `rotifer agent run` with validation checklist\n7. **Iteration** — swap underperforming Genes, tune composition\n\n## Composition Strategies\n\n| Strategy | Semantics | Use when |\n|----------|-----------|----------|\n| **Seq(A, B, C)** | Pipeline: A → B → C | Output of one feeds the next |\n| **Par(A, B)** | Parallel: run simultaneously | Independent tasks, merge results |\n| **Cond(p, A, B)** | Branch: if p then A else B | Input determines the path |\n| **Try(A, B)** | Fallback: A fails → B | Primary path unreliable |\n| **TryPool(A, B, C)** | Race: all try, first wins | Multiple equivalent implementations |\n\n## Quick Start\n\n```\nBuild a Rotifer Agent that checks document quality — grammar, readability, and tone\n```\n\nThe Skill decomposes the intent, finds top-ranked Genes, and creates the Agent.\n\n## What this Skill does on your machine\n\nIt has no code of its own — it tells your assistant which `rotifer` commands to\nrun. That is what its manifest's process-execution, filesystem and network\npermissions are for: the CLI acting, not this Skill.\n\n| | |\n|---|---|\n| **Runs** | The `rotifer` CLI (`@rotifer/playground`), fetched from npm if not installed. |\n| **Reads** | Genes and Agent definitions in the current project. |\n| **Writes** | Genes into the project's `genes/`, Agent definitions into `.rotifer/agents/`. Nothing outside the project. |\n| **Sends** | Registry and Arena queries to the public Rotifer API. Your code is not uploaded unless you run `rotifer publish` yourself. |\n\nCommands that install, publish or ov"},{"path":"_meta.json","content":"{\n \"ownerId\": \"kn7ec5m4jmk1h4jyzfrxaesjk9838ykv\",\n \"slug\": \"rotifer-agent\",\n \"version\": \"1.2.1\",\n \"publishedAt\": 1786978439199\n}"},{"path":"skill-card.md","content":"## Description:\n\nBuild a Rotifer Agent from Genes by decomposing intent into capability units, selecting Genes by Arena fitness, composing a Rotifer Genome, creating the Agent, and testing it.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[xiaoba-dev](https://clawhub.ai/user/xiaoba-dev)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and engineers use this skill to design Rotifer-specific agents by decomposing goals into Genes, selecting or creating the required Genes, choosing a Genome composition strategy, and validating the resulting Agent.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Unpinned npm execution may run a newer @rotifer/playground or @rotifer/mcp-server version than the reviewer expected.\n\nMitigation: Review or pin the npm package versions before first use.\n\nRisk: Rotifer CLI and MCP commands may read or write project-local Genes and Agent definitions.\n\nMitigation: Run the skill only in workspaces and with network access appropriate for the project data involved.\n\n## Reference(s):\n\n- [Rotifer Protocol](https://rotifer.dev)\n- [Rotifer Documentation](https://rotifer.dev/docs)\n- [Rotifer Protocol Specification](https://github.com/rotifer-protocol/rotifer-spec)\n- [Rotifer Playground](https://github.com/rotifer-protocol/rotifer-playground)\n- [ClawHub Skill Page](https://clawhub.ai/xiaoba-dev/skills/rotifer-agent)\n- [Publisher Profile](https://clawhub.ai/user/xiaoba-dev)\n\n## Skill Output:\n\n**Output Type(s):** [Guidance, Markdown, Shell commands, Configuration]\n\n**Output Format:** [Markdown with tables, checklists, JSON snippets, and inline bash commands]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May propose Rotifer CLI and MCP commands that read and write project-local Rotifer Genes and Agent definitions.]\n\n## Skill Version(s):\n\n1.2.1 (source: server release metadata and clawhub.json)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment."},{"path":"clawhub.json","content":"{\n \"name\": \"rotifer-agent\",\n \"display_name\": \"Rotifer Agent\",\n \"description\": \"Build AI Agents from Genes — decompose intent into capability units, select the fittest Genes from Arena rankings, compose them into a Rotifer Genome with Seq/Par/Cond/Try/TryPool strategies, and validate the pipeline end-to-end.\",\n \"tagline\": \"From intent to running Agent in seven steps\",\n \"version\": \"1.2.1\",\n \"author\": \"rotifer-protocol\",\n \"license\": \"Apache-2.0\",\n \"category\": \"developer-tools\",\n \"tags\": [\n \"rotifer\",\n \"ai-agent\",\n \"agent-builder\",\n \"genome\",\n \"composition\",\n \"pipeline\",\n \"gene\",\n \"orchestration\"\n ],\n \"homepage\": \"https://rotifer.dev\",\n \"repository\": \"https://github.com/rotifer-protocol/rotifer-playground\",\n \"permissions\": [\n \"process:exec\",\n \"filesystem:read\",\n \"filesystem:write\",\n \"network:outbound\"\n ],\n \"screenshots\": [],\n \"demo_video\": null\n}"}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":null,"editorialQuality":{"score":100,"threshold":65,"status":"thin","wordCount":1734,"uniquenessScore":41,"reasons":["uniqueness-below-45"]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-11T22:22:34.788Z","emptyReason":"No screenshots, media assets, or demo links are available."},"primaryImageUrl":null,"mediaAssetCount":0,"assets":[],"demoUrl":null},"ownerResources":{"evidence":{"source":"unclaimed","verified":false,"confidence":"low","updatedAt":"2026-10-11T22:22:34.788Z","emptyReason":"This page has not been claimed by the agent owner."},"hasCustomPage":false,"customPageUpdatedAt":null,"customLinks":[],"structuredLinks":{"docsUrl":null,"demoUrl":null,"supportUrl":null,"pricingUrl":null,"statusUrl":null},"customPage":null},"relatedAgents":{"evidence":{"source":"protocol-neighbors","verified":false,"confidence":"medium","updatedAt":"2026-10-12T18:25:58.434Z","emptyReason":null},"items":[{"id":"8ebccd8e-3863-4187-8355-c3f14e1f9edf","entityType":"agent","canonicalPath":"/agent/iofficeai-aionui","slug":"iofficeai-aionui","name":"AionUi","description":"Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","homepage":"https://www.aionui.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-10-09T19:11:12.944Z","createdAt":"2026-02-25T03:38:16.584Z","downloads":null},{"id":"b917f68a-ebff-438e-84f8-3f4b2494c0bc","entityType":"agent","canonicalPath":"/agent/activepieces-activepieces","slug":"activepieces-activepieces","name":"activepieces","description":"AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents","url":"https://github.com/activepieces/activepieces","homepage":"https://www.activepieces.com","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-15T02:22:12.426Z","createdAt":"2026-02-25T03:38:12.412Z","downloads":null},{"id":"5cb26759-3a39-483f-94cf-276a98c13bb8","entityType":"agent","canonicalPath":"/agent/cherryhq-cherry-studio","slug":"cherryhq-cherry-studio","name":"cherry-studio","description":"AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","homepage":"https://cherry-ai.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-11T14:38:40.986Z","createdAt":"2026-02-25T03:38:19.379Z","downloads":null},{"id":"6f6582d0-5d76-4f0f-b81d-86520247950b","entityType":"agent","canonicalPath":"/agent/copilotkit-copilotkit","slug":"copilotkit-copilotkit","name":"CopilotKit","description":"The Frontend for Agents & Generative UI. React + Angular","url":"https://github.com/CopilotKit/CopilotKit","homepage":"https://docs.copilotkit.ai","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-03-25T09:50:57.846Z","createdAt":"2026-02-25T03:39:14.617Z","downloads":null}],"links":{"hub":"/agent","source":"/agent/source/clawhub","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}