{"id":"ac42f1ae-2b34-4a4b-96a6-367f6e68491f","entityType":"agent","slug":"lzjin426-multi-agent-deliberation-skill","name":"multi-agent-deliberation","canonicalUrl":"https://www.xpersona.co/agent/lzjin426-multi-agent-deliberation-skill","canonicalPath":"/agent/lzjin426-multi-agent-deliberation-skill","generatedAt":"2026-10-09T12:59:11.862Z","source":"GITHUB_OPENCLEW","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-04-15T05:21:22.124Z","emptyReason":null},"description":"Run multi-agent deliberation when a task has arguable alternatives, testable evaluation criteria, and a falsifiable evidence chain. Use this skill to improve final answer quality and robustness via proposer/critic/checker/judge workflow with structured verdict, confidence, evidence map, residual risks, and verification steps. --- name: multi-agent-deliberation description: Run multi-agent deliberation when a task has arguable alternatives, testable evaluation criteria, and a falsifiable evidence chain. Use this skill to improve final answer quality and robustness via proposer/critic/checker/judge workflow with structured verdict, confidence, evidence map, residual risks, and verification steps. --- Multi-Agent Deliberation (v1.1) Overview","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. Last updated 4/15/2026.","installCommand":"git clone https://github.com/Lzjin426/multi-agent-deliberation-skill.git","sourceUrl":"https://github.com/Lzjin426/multi-agent-deliberation-skill","homepage":null,"primaryLinks":[{"label":"View Source","url":"https://github.com/Lzjin426/multi-agent-deliberation-skill","kind":"source"}],"safetyScore":94,"overallRank":31.3,"popularityScore":0,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Run multi-agent deliberation when a task has arguable alternatives, testable evaluation criteria, and a falsifiable evidence chain. Use this skill to improve fi"},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-04-15T05:21:22.124Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[{"label":"change","status":"self-declared"}],"verifiedCount":0,"selfDeclaredCount":2,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"},{"key":"change","type":"capability","support":"supported","confidenceSource":"profile","notes":"Declared in agent profile metadata"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile capability:change|supported|profile"}},"adoption":{"evidence":{"source":"no-adoption-signals","verified":false,"confidence":"low","updatedAt":"2026-04-15T05:21:22.124Z","emptyReason":"No source adoption metrics were available."},"stars":0,"forks":0,"downloads":null,"packageName":null,"latestVersion":null,"tractionLabel":null},"release":{"evidence":{"source":"agent-index","verified":false,"confidence":"medium","updatedAt":"2026-04-15T00:18:46.297Z","emptyReason":null},"lastUpdatedAt":"2026-04-15T05:21:22.124Z","lastCrawledAt":"2026-04-15T00:18:46.297Z","lastIndexedAt":null,"nextCrawlAt":"2026-04-16T00:18:46.297Z","lastVerifiedAt":null,"highlights":[]},"execution":{"evidence":{"source":"GITHUB OPENCLEW","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"git clone https://github.com/Lzjin426/multi-agent-deliberation-skill.git","setupComplexity":"low","setupSteps":["Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.","Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data."],"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/lzjin426-multi-agent-deliberation-skill/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/lzjin426-multi-agent-deliberation-skill/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/lzjin426-multi-agent-deliberation-skill/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/lzjin426-multi-agent-deliberation-skill/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/lzjin426-multi-agent-deliberation-skill/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/lzjin426-multi-agent-deliberation-skill/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"GITHUB_OPENCLEW","generatedAt":"2026-10-09T12:59:11.862Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/lzjin426-multi-agent-deliberation-skill/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/lzjin426-multi-agent-deliberation-skill/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/lzjin426-multi-agent-deliberation-skill/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/lzjin426-multi-agent-deliberation-skill/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":"GITHUB OPENCLEW","verified":false,"confidence":"high","updatedAt":"2026-04-15T05:21:22.124Z","emptyReason":null},"readme":"---\nname: multi-agent-deliberation\ndescription: Run multi-agent deliberation when a task has arguable alternatives, testable evaluation criteria, and a falsifiable evidence chain. Use this skill to improve final answer quality and robustness via proposer/critic/checker/judge workflow with structured verdict, confidence, evidence map, residual risks, and verification steps.\n---\n\n# Multi-Agent Deliberation (v1.1)\n\n## Overview\n\nUse this skill to run a structured multi-agent deliberation workflow for quality-first answers.\n\nPrioritize this skill only when debate adds value. Do not use it for simple factual queries or low-stakes prompts where a single agent is sufficient.\n\n## Workflow Decision Tree\n\n1. Validate activation gate.\n2. Define the decision target and testable criteria.\n3. Instantiate debate roles.\n4. Run bounded rounds with evidence constraints.\n5. Judge with fixed rubric.\n6. Emit final answer with structured state and risk controls.\n\n## Step 1: Validate Activation Gate\n\nRequire all three conditions before using multi-agent debate:\n\n1. Confirm arguable space.\n2. Confirm testable criteria.\n3. Confirm a falsifiable evidence chain.\n\nIf any condition is missing, do not spawn debate roles. Use a fallback single-agent response with `STATUS=GATE_FAILED` and list missing conditions.\n\nLoad `references/gate_and_controls.md` for the gate checklist and fallback schema.\n\n## Step 2: Define Evaluation Contract\n\nSet these items before spawning roles:\n\n1. Define `QUESTION` and explicit `CONSTRAINTS`.\n2. Define acceptance metrics.\n3. Define source quality policy.\n4. Define budget limits (rounds, token, latency).\n\nUse this default rubric unless the user specifies otherwise:\n\n- Accuracy: 40%\n- Completeness: 25%\n- Traceability: 20%\n- Robustness after rebuttal: 15%\n\nDefault source policy:\n\n- Tier A: primary sources (official docs, standards, original papers, authoritative datasets).\n- Tier B: reliable secondary synthesis.\n- Tier C: opinion/unverified content.\n- `critical claim` means a claim that can change conclusion direction, risk level, or recommended actions.\n- Critical claims require at least one Tier A source (high-stakes: two independent high-quality sources).\n\n## Step 3: Instantiate Debate Roles\n\nCreate exactly five roles by default:\n\n1. `Proposer-A`: deliver candidate answer A with evidence map.\n2. `Proposer-B`: deliver independent candidate answer B with evidence map.\n3. `Critic`: attack both candidates and expose hidden assumptions.\n4. `Evidence-Checker`: verify claim-source consistency and mark unsupported claims.\n5. `Judge`: score and decide with fixed rubric.\n\nIsolation rule:\n\n- A/B must generate first drafts in isolated contexts.\n- No cross-read of draft reasoning before Round 1 starts.\n\nLoad `references/debate_playbook.md` for reusable prompt blocks.\n\n## Step 4: Run Bounded Deliberation\n\nRun phases with explicit counting:\n\n1. Phase 0 (not counted as a debate round): independent proposals (A/B).\n2. Round 1: critic cross-examination.\n3. Round 2: targeted rebuttal and evidence updates.\n4. Optional Round 3: only if new evidence appears or core factual disagreement remains unresolved.\n\nRound budget definition:\n\n- Default debate rounds: 2 (Round 1-2).\n- Maximum debate rounds: 3.\n- Phase 0 is mandatory and excluded from round budget.\n\nStop rule:\n\n- Judge outputs total score on 0-100 scale each round.\n- `score_gain = current_total_score - previous_total_score`.\n- Stop when score gain is < 1 point for two consecutive rounds, or no new evidence appears.\n\n## Step 5: Judge and Decide\n\nForce structured decision output:\n\n1. Provide `SCORE_A` and `SCORE_B`.\n2. Select winner or merged final with explicit merge rationale.\n3. Emit final answer with confidence and residual risks.\n4. Emit what evidence would change the decision.\n5. Emit unsupported claim disposition (`dropped`, `pending_evidence`, or `accepted_with_uncertainty`).\n\nNever force consensus if evidence is weak.\n\n## Step 6: Output Contract\n\nAlways output this structure:\n\n```text\n[STATUS]\nDECIDED | MERGED | INSUFFICIENT_EVIDENCE | GATE_FAILED\n\n[FINAL_ANSWER]\n...\n\n[SCORES]\nSCORE_A: ...\nSCORE_B: ...\nSCORE_DELTA: ...\n\n[CONFIDENCE_0_TO_100]\n...\n\n[KEY_EVIDENCE_MAP]\n- claim_id -> source_id\n...\n\n[UNSUPPORTED_CLAIMS]\n- claim_id\n...\n\n[DISPOSITION]\n- claim_id: dropped | pending_evidence | accepted_with_uncertainty\n...\n\n[OPEN_RISKS]\n...\n\n[WHAT_WOULD_CHANGE_THE_ANSWER]\n...\n\n[NEXT_VERIFICATION_STEPS]\n...\n```\n\n## Operating Controls\n\n1. Cap debate rounds at 2 by default; max 3 (excluding Phase 0).\n2. Cap per-role response length to avoid verbosity bias.\n3. Require source IDs for all critical claims.\n4. Reject unsupported critical claims unless explicitly marked `accepted_with_uncertainty`.\n5. Enforce human review for all high-stakes tasks before release.\n\nLoad `references/gate_and_controls.md` for failure modes and mitigation rules.\n\n## Resource Loading Guide\n\nLoad only what is needed:\n\n1. Load `references/research_findings_2026-02-25.md` when deciding whether debate is justified and for parameter defaults.\n2. Load `references/debate_playbook.md` when you need role prompts or output templates.\n3. Load `references/gate_and_controls.md` when you need activation checks, stop rules, and risk governance.","readmeExcerpt":"--- name: multi-agent-deliberation description: Run multi-agent deliberation when a task has arguable alternatives, testable evaluation criteria, and a falsifiable evidence chain. Use this skill to improve final answer quality and robustness via proposer/critic/checker/judge workflow with structured verdict, confidence, evidence map, residual risks, and verification steps. --- Multi-Agent Deliberation (v1.1) Overview","codeSnippets":[],"executableExamples":[{"language":"text","snippet":"[STATUS]\nDECIDED | MERGED | INSUFFICIENT_EVIDENCE | GATE_FAILED\n\n[FINAL_ANSWER]\n...\n\n[SCORES]\nSCORE_A: ...\nSCORE_B: ...\nSCORE_DELTA: ...\n\n[CONFIDENCE_0_TO_100]\n...\n\n[KEY_EVIDENCE_MAP]\n- claim_id -> source_id\n...\n\n[UNSUPPORTED_CLAIMS]\n- claim_id\n...\n\n[DISPOSITION]\n- claim_id: dropped | pending_evidence | accepted_with_uncertainty\n...\n\n[OPEN_RISKS]\n...\n\n[WHAT_WOULD_CHANGE_THE_ANSWER]\n...\n\n[NEXT_VERIFICATION_STEPS]\n..."}],"parameters":{},"dependencies":[],"permissions":[],"extractedFiles":[],"languages":["typescript"],"docsSourceLabel":"GITHUB OPENCLEW","editorialOverview":"Run multi-agent deliberation when a task has arguable alternatives, testable evaluation criteria, and a falsifiable evidence chain. Use this skill to improve final answer quality and robustness via proposer/critic/checker/judge workflow with structured verdict, confidence, evidence map, residual risks, and verification steps. --- name: multi-agent-deliberation description: Run multi-agent deliberation when a task has arguable alternatives, testable evaluation criteria, and a falsifiable evidence chain. Use this skill to improve final answer quality and robustness via proposer/critic/checker/judge workflow with structured verdict, confidence, evidence map, residual risks, and verification steps. --- Multi-Agent Deliberation (v1.1) Overview","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":406,"uniquenessScore":59,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-04-15T05:21:22.124Z","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":"2026-04-15T05:21:22.124Z","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":"2026-10-09T12:59:11.862Z","emptyReason":null},"items":[{"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":"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":"2026-04-10T18:48:31.762Z","createdAt":"2026-02-25T03:38:16.584Z","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/github_openclew","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}