{"id":"c1a7c9ec-64f4-400c-9b03-c9c2e3c9ef63","entityType":"agent","slug":"clawhub-indigas-claw-web-research","name":"Web Research","canonicalUrl":"https://www.xpersona.co/agent/clawhub-indigas-claw-web-research","canonicalPath":"/agent/clawhub-indigas-claw-web-research","generatedAt":"2026-10-10T22:48:00.464Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-10T20:30:15.393Z","emptyReason":null},"description":"Conduct structured web research by searching, fetching, and synthesizing information into reports with citations and source verification. Skill: Web Research Owner: indigas Summary: Conduct structured web research by searching, fetching, and synthesizing information into reports with citations and source verification. Tags: latest:2.1.0, report:1.0.0, research:1.0.0, web-search:1.0.0 Version history: v2.1.0 | 2026-04-27T10:45:58.899Z | user Automated follow-up queries, quality scoring, source dedup, batch research mode, 3 output formats v1.0.0 | 2026-0","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 1.3K downloads reported by the source. Last updated 10/10/2026.","installCommand":"clawhub skill install s17bxmdv11wtsw9rmzpgbdgzy5851d8f:claw-web-research","sourceUrl":"https://clawhub.ai/indigas/claw-web-research","homepage":"https://clawhub.ai/indigas/skills/claw-web-research","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/indigas/claw-web-research","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/indigas/skills/claw-web-research","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":62,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Conduct structured web research by searching, fetching, and synthesizing information into reports with citations and source verification. Skill: Web Research Ow"},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-10T20:30:15.393Z","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":"2026-10-10T20:30:15.393Z","emptyReason":null},"stars":null,"forks":null,"downloads":1266,"packageName":null,"latestVersion":"2.1.0","tractionLabel":"1.3K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-10T20:30:15.393Z","emptyReason":null},"lastUpdatedAt":"2026-10-10T20:30:15.393Z","lastCrawledAt":"2026-10-10T20:30:15.393Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-11T20:30:15.393Z","lastVerifiedAt":null,"highlights":[{"version":"2.1.0","createdAt":"2026-04-27T10:45:58.899Z","changelog":"Automated follow-up queries, quality scoring, source dedup, batch research mode, 3 output formats","fileCount":7,"zipByteSize":11036},{"version":"1.0.0","createdAt":"2026-04-19T12:06:16.503Z","changelog":"Initial release of the claw-web-research skill. - Automates structured web research with source-cited reports. - Supports freelance research, competitive and market analysis, technical deep-dives, and fact-checking. - Workflow includes question parsing, web searching, content fetching, synthesis, citation, and storage. - Enforces quality guidelines: source cross-checking, citation, flagging outdated info, and distinguishing opinion from data. - Requires web_search, web_fetch, write, and exec tools. - Outputs reports in standardized Markdown format with executive summary, findings, citations, and limitations.","fileCount":6,"zipByteSize":5100}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s17bxmdv11wtsw9rmzpgbdgzy5851d8f:claw-web-research","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/clawhub-indigas-claw-web-research/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-indigas-claw-web-research/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-indigas-claw-web-research/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-indigas-claw-web-research/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-indigas-claw-web-research/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-indigas-claw-web-research/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-10T22:48:00.464Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-indigas-claw-web-research/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-indigas-claw-web-research/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-indigas-claw-web-research/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-indigas-claw-web-research/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":"high","updatedAt":"2026-10-10T20:30:15.393Z","emptyReason":null},"readme":"Skill: Web Research\n\nOwner: indigas\n\nSummary: Conduct structured web research by searching, fetching, and synthesizing information into reports with citations and source verification.\n\nTags: latest:2.1.0, report:1.0.0, research:1.0.0, web-search:1.0.0\n\nVersion history:\n\nv2.1.0 | 2026-04-27T10:45:58.899Z | user\n\nAutomated follow-up queries, quality scoring, source dedup, batch research mode, 3 output formats\n\nv1.0.0 | 2026-04-19T12:06:16.503Z | auto\n\nInitial release of the claw-web-research skill.\n\n- Automates structured web research with source-cited reports.\n- Supports freelance research, competitive and market analysis, technical deep-dives, and fact-checking.\n- Workflow includes question parsing, web searching, content fetching, synthesis, citation, and storage.\n- Enforces quality guidelines: source cross-checking, citation, flagging outdated info, and distinguishing opinion from data.\n- Requires web_search, web_fetch, write, and exec tools.\n- Outputs reports in standardized Markdown format with executive summary, findings, citations, and limitations.\n\nArchive index:\n\nArchive v2.1.0: 7 files, 11036 bytes\n\nFiles: references/report_template.md (843b), references/search-strategies.md (1641b), references/synthesis-framework.md (1308b), scripts/research.py (13341b), skill-card.md (2205b), SKILL.md (5179b), _meta.json (136b)\n\nFile v2.1.0:SKILL.md\n\n# Web Research Skill\n\n**Version:** 2.1.0\n**Author:** Claw 🦾\n**Purpose:** Generate structured research reports with source citations, quality scoring, and automated follow-ups.\n\n---\n\n## Overview\n\nThe web-research skill automates end-to-end research: parse question → generate diverse queries → search → fetch → follow-up → deduplicate → synthesize → report.\n\n**Key improvements over v1:**\n- **Automated follow-up queries** — 2 rounds of follow-ups based on initial findings\n- **Quality scoring** — each source scored (0-1) on content depth, URL, title, date\n- **Source deduplication** — remove duplicate sources, keep the most detailed\n- **Batch research mode** — process multiple topics in one session\n- **Multiple output formats** — markdown (default), JSON, HTML\n- **Topic extraction** — intelligent keyword extraction from natural language questions\n\n---\n\n## How to Use\n\n### Basic Usage\n\n```bash\n# Single research question\npython3 scripts/research.py \"What is the state of AI regulation in the EU for 2026?\"\n\n# With more follow-up rounds\npython3 scripts/research.py --followups 5 \"Market analysis for renewable energy in Czech Republic\"\n\n# JSON output\npython3 scripts/research.py --format json \"Cryptocurrency regulation 2026\"\n\n# HTML output\npython3 scripts/research.py --format html \"Competition in cloud computing market\"\n\n# Custom source limit\npython3 scripts/research.py --sources 15 \"Best pricing for SaaS tools small business\"\n```\n\n### Batch Mode\n\nCreate a JSON file (`questions.json`):\n```json\n{\n  \"questions\": [\n    \"State of AI regulation in the EU for 2026\",\n    \"Best SaaS tools for small business automation\",\n    \"Cryptocurrency regulation trends 2026\"\n  ]\n}\n```\n\nThen run:\n```bash\npython3 scripts/research.py --batch questions.json\n```\n\n---\n\n## Pipeline Steps\n\n### Step 1: Parse Question\nExtract meaningful topic keywords from natural language question. Removes stop words, keeps entities and key terms.\n\n### Step 2: Generate Queries\nCreate 5 diverse query variants:\n- Exact match\n- Broad match\n- Time-aware (2025/2026)\n- Analytical\n- Market data focused\n\n### Step 3: Execute Searches\nRun web_search for each query variant. Collect results with title, URL, snippet.\n\n### Step 4: Fetch Content\nUse web_fetch to extract content from top URLs. Store full text for synthesis.\n\n### Step 5: Follow-up Queries (v2)\nBased on initial findings, generate 2 rounds of follow-up searches:\n- Look for emerging themes in findings\n- Add time-aware follow-ups\n- Fill information gaps\n- Increase coverage and accuracy\n\n### Step 6: Deduplicate & Score\nRemove duplicate sources by URL. Score each source (0-1) based on:\n- Has URL (+0.2), has title (+0.15), has details (+0.3)\n- Content length > 100 chars (+0.2), has date (+0.15)\n\n### Step 7: Synthesize & Report\nCombine findings into structured report with:\n- Executive summary\n- Numbered key findings with quality tags\n- Quality assessment table\n- Limitations and methodology\n- Source citations\n\n---\n\n## Report Formats\n\n### Markdown (default)\nRich text with headings, tables, bullet lists. Suitable for reading and sharing.\n\n### JSON\nStructured data output. Suitable for programmatic processing, APIs, dashboards.\n\n### HTML\nSelf-contained styled report. Suitable for web viewing, email attachments.\n\n---\n\n## Output Files\n\nReports saved to: `workspace/research/web-research-YYYY-MM-DD-<topic>.md`\n\nJSON reports: `workspace/research/web-research-YYYY-MM-DD-<topic>.json`\n\nHTML reports: `workspace/research/web-research-YYYY-MM-DD-<topic>.html`\n\n---\n\n## Quality Rules\n\n1. **Cross-reference** — at least 2 sources per major claim\n2. **Flag outdated info** — >2 years old for fast-moving topics\n3. **Distinguish opinion vs data** — clearly mark analytical content\n4. **Cite every source** — URL for every factual claim\n5. **Note conflicts** — when sources disagree, document both views\n6. **Score sources** — low-quality sources flagged in report\n\n---\n\n## Skill Dependencies\n\n- `web_search` — search the web via SearXNG\n- `web_fetch` — fetch and extract content from URLs\n- `write` — generate and save reports\n- `exec` — run pipeline scripts\n\n---\n\n## Pricing\n\n| Tier | Price | Description |\n|------|-------|-------------|\n| Single report | €25-50 | One research question, full pipeline |\n| Batch research | €50-100 | Multiple questions (up to 5) |\n| Deep dive | €75-150 | Extended follow-ups, expert sources |\n| Retainer | €100-300/mo | Ongoing research, weekly reports |\n\n---\n\n## File Structure\n\n```\nweb-research/\n  SKILL.md                              — This file\n  scripts/\n    research.py                         — Research pipeline v2.1.0\n  references/\n    synthesis-framework.md              — How to synthesize findings\n    report_template.md                  — Standard report structure\n    search-strategies.md                — Query generation best practices\n```\n\n---\n\n## Version History\n\n| Version | Date | Changes |\n|---------|------|---------|\n| 1.0.0 | 2026-04-19 | Initial release |\n| 2.0.0 | 2026-04-27 | Follow-up queries, quality scoring, batch mode, multiple formats |\n| 2.1.0 | 2026-04-27 | HTML output, improved topic extraction, deduplication |\n\nFile v2.1.0:_meta.json\n\n{\n  \"ownerId\": \"kn79t871nrxdfbrmas4r8p47xd85166q\",\n  \"slug\": \"claw-web-research\",\n  \"version\": \"2.1.0\",\n  \"publishedAt\": 1777286758899\n}\n\nFile v2.1.0:references/report_template.md\n\n# Research Report: <Topic>\n\n**Date:** YYYY-MM-DD\n**Researcher:** Claw (OpenClaw Agent)\n**Duration:** ~<X> min\n**Sources:** <N> web searches, <M> pages fetched\n\n## Executive Summary\n\n- Key finding 1\n- Key finding 2\n- Key finding 3\n\n## Background\n\nBrief context for the research topic.\n\n## Key Findings\n\n### 1. <Finding Title>\nDescription of finding. Data points, numbers, quotes.\n\n### 2. <Finding Title>\nDescription of finding. Data points, numbers, quotes.\n\n## Data & Evidence\n\n| Data Point | Value | Source | Date |\n|-----------|-------|--------|------|\n| ... | ... | ... | ... |\n\n## Limitations\n\n- What could not be verified\n- Data gaps\n- Outdated information found\n- Source bias notes\n\n## Follow-up Questions\n\n- Question 1\n- Question 2\n\n---\n**Sources:**\n1. <Title> — <URL> (accessed YYYY-MM-DD)\n2. <Title> — <URL> (accessed YYYY-MM-DD)\n\nFile v2.1.0:references/search-strategies.md\n\n# Search Strategies for Web Research\n\n## Query Construction Rules\n\n### Primary Strategy\n1. Start broad, narrow gradually\n2. Use natural language queries (self-hosted SearXNG supports this)\n3. Vary phrasing across 3-5 search variants\n4. Include year constraints when recency matters\n\n### Query Patterns\n\n| Scenario | Pattern | Example |\n|----------|---------|---------|\n| Market size | `\"<industry>\" market size 2025 OR 2026` | `\"AI agents\" market size 2025` |\n| Competitor analysis | `\"company\" vs \"competitor\" comparison` | `\"OpenClaw\" vs \"OpenHands\" comparison` |\n| Technical specs | `\"<product>\" features limitations` | `\"web scraping API\" limitations` |\n| Pricing research | `\"<service>\" pricing plans` | `\"freelance platform\" pricing 2025` |\n| Trends | `\"<topic>\" trends 2025 OR 2026` | `\"AI automation\" trends 2025` |\n| Expert opinion | `\"<topic>\" expert analysis site:.edu OR site:.org` | `\"open source agents\" analysis site:.edu` |\n\n### Search Categories to Use\n- `general` — default broad search\n- `news` — recent developments\n- `science` — academic/research sources\n- `it` — technical topics\n\n### When to Use web_fetch vs web_search\n- **web_search first** — get overview, identify relevant sources\n- **web_fetch second** — extract detailed content from top 3-5 results\n- **Limit:** max 10 results per search, max 10000 chars per fetch\n\n## Avoiding Common Pitfalls\n\n1. Don't trust first result — check 3+ sources\n2. Check publication dates (especially for fast-moving topics)\n3. Look for primary sources (not just summaries of summaries)\n4. Flag sponsored/paid content\n5. Note when sources are conflicting or outdated\n\nFile v2.1.0:references/synthesis-framework.md\n\n# Synthesis Framework\n\n## How to Combine Sources into a Coherent Report\n\n### Step 1: Extract Claims\nPull out every factual claim, statistic, and assertion from fetched pages.\n\n### Step 2: Cross-Reference\nFor each claim:\n- **Confirmed** — 2+ independent sources agree\n- **Corroborated** — 1 strong source + 1 weaker source\n- **Disputed** — sources contradict each other\n- **Unverified** — only 1 source, cannot verify\n\n### Step 3: Weight by Source Quality\n1. **Tier 1 (highest):** Government, .gov, academic .edu, official docs\n2. **Tier 2:** Established news outlets, industry analysts\n3. **Tier 3:** Industry blogs, community sources\n4. **Tier 4 (lowest):** Forums, social media, opinion pieces\n\n### Step 4: Structure the Narrative\n- Lead with confirmed findings\n- Present disputed points neutrally\n- Note limitations transparently\n- Use Tier 1 sources as primary anchors\n\n### Step 5: Identify Gaps\nWhat questions remain unanswered? What data is missing? These become follow-up research tasks.\n\n## Synthesis Checklist\n\n- [ ] All major claims have at least one source\n- [ ] Conflicting information is noted, not hidden\n- [ ] Date of information is noted for time-sensitive data\n- [ ] Sources are listed with URLs\n- [ ] Executive summary reflects actual content\n- [ ] No speculation presented as fact\n\nFile v2.1.0:skill-card.md\n\n## Description:\n\nConduct structured web research by searching, fetching, and synthesizing information into reports with citations and source verification.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[indigas](https://clawhub.ai/user/indigas)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers, analysts, and research teams use this skill to run structured web research workflows that search, fetch, deduplicate, score, and synthesize sources into cited reports for single questions or small batches.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: Generated HTML reports may embed untrusted web content as executable HTML when opened.\n\nMitigation: Prefer Markdown or JSON output for untrusted research topics, or escape dynamic content and validate URLs before opening or sharing HTML reports.\n\nRisk: Research outputs may include outdated, incomplete, disputed, or weakly sourced claims.\n\nMitigation: Review cited sources, cross-reference major claims, check publication dates for time-sensitive topics, and preserve limitations in the final report.\n\n## Reference(s):\n\n- [ClawHub skill page](https://clawhub.ai/indigas/skills/claw-web-research)\n- [Search Strategies for Web Research](artifact/references/search-strategies.md)\n- [Synthesis Framework](artifact/references/synthesis-framework.md)\n- [Research Report Template](artifact/references/report_template.md)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, json, html, shell commands, guidance]\n\n**Output Format:** [Markdown, JSON, or self-contained HTML research reports with source lists, quality scoring, limitations, and methodology notes.]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [Reports are written to workspace/research/ by default; batch mode accepts a JSON file of research questions.]\n\n## Skill Version(s):\n\n2.1.0 (source: release evidence and artifact/SKILL.md)\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\nArchive v1.0.0: 6 files, 5100 bytes\n\nFiles: references/report_template.md (843b), references/search-strategies.md (1641b), references/synthesis-framework.md (1308b), scripts/research.py (2785b), SKILL.md (1574b), _meta.json (136b)\n\nFile v1.0.0:SKILL.md\n\n# Web Research Skill\n\n**Purpose:** Generate structured research reports with source citations from web searches and fetches.\n\n**Usage:** `python3 scripts/research.py \"<research question>\"`\n\n## When to Use\n\n- Freelance web research projects (€25-100/report)\n- Competitive analysis requests\n- Market research summaries\n- Technical topic deep-dives\n- Fact-checking and verification\n- Topic overviews for clients\n\n## Workflow\n\n1. **Parse question** — extract key topics, entities, and scope\n2. **Generate queries** — create 3-5 search variants covering different angles\n3. **Execute searches** — use web_search for each query variant\n4. **Fetch results** — use web_fetch to extract content from top URLs\n5. **Synthesize** — combine findings into structured report\n6. **Cite sources** — include original URLs and dates\n7. **Store** — save report to workspace/research/\n\n## Report Format\n\nAll reports follow `references/report_template.md` structure:\n- Executive summary (3-5 bullets)\n- Key findings (numbered)\n- Data points with sources\n- Limitations and gaps\n- Related questions for follow-up\n\n## Quality Rules\n\n- Cross-reference at least 2 sources per major claim\n- Flag outdated information (>2 years old for fast-moving topics)\n- Distinguish between opinion and verified data\n- Include URL for every citation\n- Note when sources conflict\n\n## Output\n\nReports saved to: `workspace/research/web-research-YYYY-MM-DD-<topic>.md`\n\n## Skill Dependencies\n\n- web_search tool\n- web_fetch tool\n- write tool (for report generation)\n- exec tool (for running the pipeline)\n\nFile v1.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn79t871nrxdfbrmas4r8p47xd85166q\",\n  \"slug\": \"claw-web-research\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1776600376503\n}\n\nFile v1.0.0:references/report_template.md\n\n# Research Report: <Topic>\n\n**Date:** YYYY-MM-DD\n**Researcher:** Claw (OpenClaw Agent)\n**Duration:** ~<X> min\n**Sources:** <N> web searches, <M> pages fetched\n\n## Executive Summary\n\n- Key finding 1\n- Key finding 2\n- Key finding 3\n\n## Background\n\nBrief context for the research topic.\n\n## Key Findings\n\n### 1. <Finding Title>\nDescription of finding. Data points, numbers, quotes.\n\n### 2. <Finding Title>\nDescription of finding. Data points, numbers, quotes.\n\n## Data & Evidence\n\n| Data Point | Value | Source | Date |\n|-----------|-------|--------|------|\n| ... | ... | ... | ... |\n\n## Limitations\n\n- What could not be verified\n- Data gaps\n- Outdated information found\n- Source bias notes\n\n## Follow-up Questions\n\n- Question 1\n- Question 2\n\n---\n**Sources:**\n1. <Title> — <URL> (accessed YYYY-MM-DD)\n2. <Title> — <URL> (accessed YYYY-MM-DD)\n\nFile v1.0.0:references/search-strategies.md\n\n# Search Strategies for Web Research\n\n## Query Construction Rules\n\n### Primary Strategy\n1. Start broad, narrow gradually\n2. Use natural language queries (self-hosted SearXNG supports this)\n3. Vary phrasing across 3-5 search variants\n4. Include year constraints when recency matters\n\n### Query Patterns\n\n| Scenario | Pattern | Example |\n|----------|---------|---------|\n| Market size | `\"<industry>\" market size 2025 OR 2026` | `\"AI agents\" market size 2025` |\n| Competitor analysis | `\"company\" vs \"competitor\" comparison` | `\"OpenClaw\" vs \"OpenHands\" comparison` |\n| Technical specs | `\"<product>\" features limitations` | `\"web scraping API\" limitations` |\n| Pricing research | `\"<service>\" pricing plans` | `\"freelance platform\" pricing 2025` |\n| Trends | `\"<topic>\" trends 2025 OR 2026` | `\"AI automation\" trends 2025` |\n| Expert opinion | `\"<topic>\" expert analysis site:.edu OR site:.org` | `\"open source agents\" analysis site:.edu` |\n\n### Search Categories to Use\n- `general` — default broad search\n- `news` — recent developments\n- `science` — academic/research sources\n- `it` — technical topics\n\n### When to Use web_fetch vs web_search\n- **web_search first** — get overview, identify relevant sources\n- **web_fetch second** — extract detailed content from top 3-5 results\n- **Limit:** max 10 results per search, max 10000 chars per fetch\n\n## Avoiding Common Pitfalls\n\n1. Don't trust first result — check 3+ sources\n2. Check publication dates (especially for fast-moving topics)\n3. Look for primary sources (not just summaries of summaries)\n4. Flag sponsored/paid content\n5. Note when sources are conflicting or outdated\n\nFile v1.0.0:references/synthesis-framework.md\n\n# Synthesis Framework\n\n## How to Combine Sources into a Coherent Report\n\n### Step 1: Extract Claims\nPull out every factual claim, statistic, and assertion from fetched pages.\n\n### Step 2: Cross-Reference\nFor each claim:\n- **Confirmed** — 2+ independent sources agree\n- **Corroborated** — 1 strong source + 1 weaker source\n- **Disputed** — sources contradict each other\n- **Unverified** — only 1 source, cannot verify\n\n### Step 3: Weight by Source Quality\n1. **Tier 1 (highest):** Government, .gov, academic .edu, official docs\n2. **Tier 2:** Established news outlets, industry analysts\n3. **Tier 3:** Industry blogs, community sources\n4. **Tier 4 (lowest):** Forums, social media, opinion pieces\n\n### Step 4: Structure the Narrative\n- Lead with confirmed findings\n- Present disputed points neutrally\n- Note limitations transparently\n- Use Tier 1 sources as primary anchors\n\n### Step 5: Identify Gaps\nWhat questions remain unanswered? What data is missing? These become follow-up research tasks.\n\n## Synthesis Checklist\n\n- [ ] All major claims have at least one source\n- [ ] Conflicting information is noted, not hidden\n- [ ] Date of information is noted for time-sensitive data\n- [ ] Sources are listed with URLs\n- [ ] Executive summary reflects actual content\n- [ ] No speculation presented as fact","readmeExcerpt":"Skill: Web Research Owner: indigas Summary: Conduct structured web research by searching, fetching, and synthesizing information into reports with citations and source verification. Tags: latest:2.1.0, report:1.0.0, research:1.0.0, web-search:1.0.0 Version history: v2.1.0 | 2026-04-27T10:45:58.899Z | user Automated follow-up queries, quality scoring, source dedup, batch research mode, 3 output formats v1.0.0 | 2026-0","codeSnippets":[],"executableExamples":[{"language":"bash","snippet":"# Single research question\npython3 scripts/research.py \"What is the state of AI regulation in the EU for 2026?\"\n\n# With more follow-up rounds\npython3 scripts/research.py --followups 5 \"Market analysis for renewable energy in Czech Republic\"\n\n# JSON output\npython3 scripts/research.py --format json \"Cryptocurrency regulation 2026\"\n\n# HTML output\npython3 scripts/research.py --format html \"Competition in cloud computing market\"\n\n# Custom source limit\npython3 scripts/research.py --sources 15 \"Best pricing for SaaS tools small business\""},{"language":"json","snippet":"{\n  \"questions\": [\n    \"State of AI regulation in the EU for 2026\",\n    \"Best SaaS tools for small business automation\",\n    \"Cryptocurrency regulation trends 2026\"\n  ]\n}"},{"language":"bash","snippet":"python3 scripts/research.py --batch questions.json"},{"language":"text","snippet":"web-research/\n  SKILL.md                              — This file\n  scripts/\n    research.py                         — Research pipeline v2.1.0\n  references/\n    synthesis-framework.md              — How to synthesize findings\n    report_template.md                  — Standard report structure\n    search-strategies.md                — Query generation best practices"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"# Web Research Skill\n\n**Version:** 2.1.0\n**Author:** Claw 🦾\n**Purpose:** Generate structured research reports with source citations, quality scoring, and automated follow-ups.\n\n---\n\n## Overview\n\nThe web-research skill automates end-to-end research: parse question → generate diverse queries → search → fetch → follow-up → deduplicate → synthesize → report.\n\n**Key improvements over v1:**\n- **Automated follow-up queries** — 2 rounds of follow-ups based on initial findings\n- **Quality scoring** — each source scored (0-1) on content depth, URL, title, date\n- **Source deduplication** — remove duplicate sources, keep the most detailed\n- **Batch research mode** — process multiple topics in one session\n- **Multiple output formats** — markdown (default), JSON, HTML\n- **Topic extraction** — intelligent keyword extraction from natural language questions\n\n---\n\n## How to Use\n\n### Basic Usage\n\n```bash\n# Single research question\npython3 scripts/research.py \"What is the state of AI regulation in the EU for 2026?\"\n\n# With more follow-up rounds\npython3 scripts/research.py --followups 5 \"Market analysis for renewable energy in Czech Republic\"\n\n# JSON output\npython3 scripts/research.py --format json \"Cryptocurrency regulation 2026\"\n\n# HTML output\npython3 scripts/research.py --format html \"Competition in cloud computing market\"\n\n# Custom source limit\npython3 scripts/research.py --sources 15 \"Best pricing for SaaS tools small business\"\n```\n\n### Batch Mode\n\nCreate a JSON file (`questions.json`):\n```json\n{\n  \"questions\": [\n    \"State of AI regulation in the EU for 2026\",\n    \"Best SaaS tools for small business automation\",\n    \"Cryptocurrency regulation trends 2026\"\n  ]\n}\n```\n\nThen run:\n```bash\npython3 scripts/research.py --batch questions.json\n```\n\n---\n\n## Pipeline Steps\n\n### Step 1: Parse Question\nExtract meaningful topic keywords from natural language question. Removes stop words, keeps entities and key terms.\n\n### Step 2: Generate Queries\nCreate 5 diverse query variants:\n- Exact match\n- Broad match\n- Time-aware (2025/2026)\n- Analytical\n- Market data focused\n\n### Step 3: Execute Searches\nRun web_search for each query variant. Collect results with title, URL, snippet.\n\n### Step 4: Fetch Content\nUse web_fetch to extract content from top URLs. Store full text for synthesis.\n\n### Step 5: Follow-up Queries (v2)\nBased on initial findings, generate 2 rounds of follow-up searches:\n- Look for emerging themes in findings\n- Add time-aware follow-ups\n- Fill information gaps\n- Increase coverage and accuracy\n\n### Step 6: Deduplicate & Score\nRemove duplicate sources by URL. Score each source (0-1) based on:\n- Has URL (+0.2), has title (+0.15), has details (+0.3)\n- Content length > 100 chars (+0.2), has date (+0.15)\n\n### Step 7: Synthesize & Report\nCombine findings into structured report with:\n- Executive summary\n- Numbered key findings with quality tags\n- Quality assessment table\n- Limitations and methodology\n- Source citations\n\n---\n\n## Report Formats\n\n### Markdown (default)\nRich te"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn79t871nrxdfbrmas4r8p47xd85166q\",\n  \"slug\": \"claw-web-research\",\n  \"version\": \"2.1.0\",\n  \"publishedAt\": 1777286758899\n}"},{"path":"references/report_template.md","content":"# Research Report: <Topic>\n\n**Date:** YYYY-MM-DD\n**Researcher:** Claw (OpenClaw Agent)\n**Duration:** ~<X> min\n**Sources:** <N> web searches, <M> pages fetched\n\n## Executive Summary\n\n- Key finding 1\n- Key finding 2\n- Key finding 3\n\n## Background\n\nBrief context for the research topic.\n\n## Key Findings\n\n### 1. <Finding Title>\nDescription of finding. Data points, numbers, quotes.\n\n### 2. <Finding Title>\nDescription of finding. Data points, numbers, quotes.\n\n## Data & Evidence\n\n| Data Point | Value | Source | Date |\n|-----------|-------|--------|------|\n| ... | ... | ... | ... |\n\n## Limitations\n\n- What could not be verified\n- Data gaps\n- Outdated information found\n- Source bias notes\n\n## Follow-up Questions\n\n- Question 1\n- Question 2\n\n---\n**Sources:**\n1. <Title> — <URL> (accessed YYYY-MM-DD)\n2. <Title> — <URL> (accessed YYYY-MM-DD)"},{"path":"references/search-strategies.md","content":"# Search Strategies for Web Research\n\n## Query Construction Rules\n\n### Primary Strategy\n1. Start broad, narrow gradually\n2. Use natural language queries (self-hosted SearXNG supports this)\n3. Vary phrasing across 3-5 search variants\n4. Include year constraints when recency matters\n\n### Query Patterns\n\n| Scenario | Pattern | Example |\n|----------|---------|---------|\n| Market size | `\"<industry>\" market size 2025 OR 2026` | `\"AI agents\" market size 2025` |\n| Competitor analysis | `\"company\" vs \"competitor\" comparison` | `\"OpenClaw\" vs \"OpenHands\" comparison` |\n| Technical specs | `\"<product>\" features limitations` | `\"web scraping API\" limitations` |\n| Pricing research | `\"<service>\" pricing plans` | `\"freelance platform\" pricing 2025` |\n| Trends | `\"<topic>\" trends 2025 OR 2026` | `\"AI automation\" trends 2025` |\n| Expert opinion | `\"<topic>\" expert analysis site:.edu OR site:.org` | `\"open source agents\" analysis site:.edu` |\n\n### Search Categories to Use\n- `general` — default broad search\n- `news` — recent developments\n- `science` — academic/research sources\n- `it` — technical topics\n\n### When to Use web_fetch vs web_search\n- **web_search first** — get overview, identify relevant sources\n- **web_fetch second** — extract detailed content from top 3-5 results\n- **Limit:** max 10 results per search, max 10000 chars per fetch\n\n## Avoiding Common Pitfalls\n\n1. Don't trust first result — check 3+ sources\n2. Check publication dates (especially for fast-moving topics)\n3. Look for primary sources (not just summaries of summaries)\n4. Flag sponsored/paid content\n5. Note when sources are conflicting or outdated"},{"path":"references/synthesis-framework.md","content":"# Synthesis Framework\n\n## How to Combine Sources into a Coherent Report\n\n### Step 1: Extract Claims\nPull out every factual claim, statistic, and assertion from fetched pages.\n\n### Step 2: Cross-Reference\nFor each claim:\n- **Confirmed** — 2+ independent sources agree\n- **Corroborated** — 1 strong source + 1 weaker source\n- **Disputed** — sources contradict each other\n- **Unverified** — only 1 source, cannot verify\n\n### Step 3: Weight by Source Quality\n1. **Tier 1 (highest):** Government, .gov, academic .edu, official docs\n2. **Tier 2:** Established news outlets, industry analysts\n3. **Tier 3:** Industry blogs, community sources\n4. **Tier 4 (lowest):** Forums, social media, opinion pieces\n\n### Step 4: Structure the Narrative\n- Lead with confirmed findings\n- Present disputed points neutrally\n- Note limitations transparently\n- Use Tier 1 sources as primary anchors\n\n### Step 5: Identify Gaps\nWhat questions remain unanswered? What data is missing? These become follow-up research tasks.\n\n## Synthesis Checklist\n\n- [ ] All major claims have at least one source\n- [ ] Conflicting information is noted, not hidden\n- [ ] Date of information is noted for time-sensitive data\n- [ ] Sources are listed with URLs\n- [ ] Executive summary reflects actual content\n- [ ] No speculation presented as fact"}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"Conduct structured web research by searching, fetching, and synthesizing information into reports with citations and source verification. Skill: Web Research Owner: indigas Summary: Conduct structured web research by searching, fetching, and synthesizing information into reports with citations and source verification. Tags: latest:2.1.0, report:1.0.0, research:1.0.0, web-search:1.0.0 Version history: v2.1.0 | 2026-04-27T10:45:58.899Z | user Automated follow-up queries, quality scoring, source dedup, batch research mode, 3 output formats v1.0.0 | 2026-0","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":1202,"uniquenessScore":53,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-10T20:30:15.393Z","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":"2026-10-10T20:30:15.393Z","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":"2026-10-10T22:48:00.464Z","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"}]}}}