{"id":"d3b55db3-c5c1-4109-9c0e-a34a8e9d50d8","entityType":"agent","slug":"clawhub-wd041216-bit-openclaw-free-web-search","name":"Zero Api Key Web Search","canonicalUrl":"https://www.xpersona.co/agent/clawhub-wd041216-bit-openclaw-free-web-search","canonicalPath":"/agent/clawhub-wd041216-bit-openclaw-free-web-search","generatedAt":"2026-10-10T08:09:53.634Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-09T07:56:24.950Z","emptyReason":null},"description":"OpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking. No API keys required by default; optional providers can be enab... Skill: Zero Api Key Web Search Owner: wd041216-bit Summary: OpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking. No API keys required by default; optional providers can be enab... Tags: fact-check:21.0.0, latest:23.0.0, mcp:21.0.0, no-api-key:21.0.0, openclaw:21.0.0, v11:11.0.0, verification:21.0.0, web-search:21.0.0 Version history: v23.0.0 | 2026-05-03T08:52:33.463Z | user v","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 3.5K downloads reported by the source. Last updated 10/9/2026.","installCommand":"clawhub skill install s17djcs4rd60zq6rxnpn9xv3d583g4ew:openclaw-free-web-search","sourceUrl":"https://clawhub.ai/wd041216-bit/openclaw-free-web-search","homepage":"https://clawhub.ai/wd041216-bit/skills/openclaw-free-web-search","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/wd041216-bit/openclaw-free-web-search","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/wd041216-bit/skills/openclaw-free-web-search","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":61,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"OpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking. No API keys required by default; optional providers can be enab..."},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-09T07:56:24.950Z","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":"2026-10-09T07:56:24.950Z","emptyReason":null},"stars":null,"forks":null,"downloads":3475,"packageName":null,"latestVersion":"23.0.0","tractionLabel":"3.5K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-09T07:56:24.950Z","emptyReason":null},"lastUpdatedAt":"2026-10-09T07:56:24.950Z","lastCrawledAt":"2026-10-09T07:56:24.950Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-10T07:56:24.950Z","lastVerifiedAt":null,"highlights":[{"version":"23.0.0","createdAt":"2026-05-03T08:52:33.463Z","changelog":"v23: multi-engine SERP (7 engines via Bright Data), Web Unlocker for blocked pages with auto-fallback on 403/429, production-unlock profile, MCP server engine param","fileCount":3,"zipByteSize":3467},{"version":"21.0.0","createdAt":"2026-05-01T00:39:57.068Z","changelog":"v21: provider profiles (free/free-verified/production/max-evidence), LLM context mode via zero-context, Goggles-lite reranking (docs-first/research/news-balanced), agent search controls, Bright Data provider, semantic oracle boundary, claim ledger, evaluation framework, npm wrapper published","fileCount":2,"zipByteSize":1476},{"version":"18.0.0","createdAt":"2026-04-19T22:07:43.611Z","changelog":"Version 18.0.0 - Renamed skill to \"zero-api-key-web-search\" with improved, platform-agnostic branding. - No API keys required—100% free search, browsing, and claim verification. - Introduced new CLI commands: zero-search, zero-browse, zero-verify, zero-report (legacy aliases retained). - Expanded compatibility and setup details for multiple AI agent platforms. - Added comprehensive usage guidance, compatibility notes, and updated documentation (README.md, CHANGELOG.md). - Emphasized evidence surfacing, citation, and free dual-provider strategy.","fileCount":4,"zipByteSize":8638},{"version":"16.0.0","createdAt":"2026-03-24T02:07:57.011Z","changelog":"Refresh legacy OpenClaw skill slug to point to the renamed cross-validated-search project with evidence-report, verify-claim, and free ddgs + self-hosted SearXNG guidance.","fileCount":2,"zipByteSize":1324},{"version":"11.0.0","createdAt":"2026-03-17T19:00:17.644Z","changelog":"Free Web Search Ultimate v11.0 adopts a \"Search-First\" paradigm: - Major refactor: Replaces previous scripts and docs with a streamlined, unified interface (`search-web`, `browse-page`) and guidelines for agents. - Provides concise behavior guidelines and a clear decision tree to enforce searching live web before answering factual or timely queries. - Simplifies command structure for web, news, images, books, and videos search, with improved region and timeframe controls. - Details universal integration (OpenClaw, MCP server, LangChain, OpenAI) for LLM agents and pip install instructions. - Removes older implementation/agent/script files and replaces existing SKILL.md with comprehensive new documentation focused on agent-centric behavior. - Stronger emphasis on real-time accuracy, cite-every-fact policy, and privacy-first design.","fileCount":2,"zipByteSize":2279},{"version":"4.0.0","createdAt":"2026-03-13T12:06:19.308Z","changelog":"v4.0: Added verify_claim.py for multi-source cross-validation. Now tells you how much to trust the answer with VERIFIED/LIKELY_TRUE/UNCERTAIN/LIKELY_FALSE verdicts. Multi-engine parallel search + 4-tier anti-bot cascade + authority-weighted scoring.","fileCount":5,"zipByteSize":16322},{"version":"1.0.0","createdAt":"2026-03-13T11:19:36.365Z","changelog":"Initial release - Free, private web search for OpenClaw with self-hosted SearXNG + Scrapling anti-bot fetcher. Zero API keys, zero cost.","fileCount":5,"zipByteSize":16321}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s17djcs4rd60zq6rxnpn9xv3d583g4ew:openclaw-free-web-search","setupComplexity":"low","setupSteps":["Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.","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-wd041216-bit-openclaw-free-web-search/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-wd041216-bit-openclaw-free-web-search/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-wd041216-bit-openclaw-free-web-search/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-wd041216-bit-openclaw-free-web-search/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-wd041216-bit-openclaw-free-web-search/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-wd041216-bit-openclaw-free-web-search/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-10T08:09:53.633Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-wd041216-bit-openclaw-free-web-search/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-wd041216-bit-openclaw-free-web-search/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-wd041216-bit-openclaw-free-web-search/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-wd041216-bit-openclaw-free-web-search/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":"2026-10-09T07:56:24.950Z","emptyReason":null},"readme":"Skill: Zero Api Key Web Search\n\nOwner: wd041216-bit\n\nSummary: OpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking. No API keys required by default; optional providers can be enab...\n\nTags: fact-check:21.0.0, latest:23.0.0, mcp:21.0.0, no-api-key:21.0.0, openclaw:21.0.0, v11:11.0.0, verification:21.0.0, web-search:21.0.0\n\nVersion history:\n\nv23.0.0 | 2026-05-03T08:52:33.463Z | user\n\nv23: multi-engine SERP (7 engines via Bright Data), Web Unlocker for blocked pages with auto-fallback on 403/429, production-unlock profile, MCP server engine param\n\nv21.0.0 | 2026-05-01T00:39:57.068Z | user\n\nv21: provider profiles (free/free-verified/production/max-evidence), LLM context mode via zero-context, Goggles-lite reranking (docs-first/research/news-balanced), agent search controls, Bright Data provider, semantic oracle boundary, claim ledger, evaluation framework, npm wrapper published\n\nv18.0.0 | 2026-04-19T22:07:43.611Z | user\n\nVersion 18.0.0\n\n- Renamed skill to \"zero-api-key-web-search\" with improved, platform-agnostic branding.\n- No API keys required—100% free search, browsing, and claim verification.\n- Introduced new CLI commands: zero-search, zero-browse, zero-verify, zero-report (legacy aliases retained).\n- Expanded compatibility and setup details for multiple AI agent platforms.\n- Added comprehensive usage guidance, compatibility notes, and updated documentation (README.md, CHANGELOG.md).\n- Emphasized evidence surfacing, citation, and free dual-provider strategy.\n\nv16.0.0 | 2026-03-24T02:07:57.011Z | user\n\nRefresh legacy OpenClaw skill slug to point to the renamed cross-validated-search project with evidence-report, verify-claim, and free ddgs + self-hosted SearXNG guidance.\n\nv11.0.0 | 2026-03-17T19:00:17.644Z | auto\n\nFree Web Search Ultimate v11.0 adopts a \"Search-First\" paradigm:\n\n- Major refactor: Replaces previous scripts and docs with a streamlined, unified interface (`search-web`, `browse-page`) and guidelines for agents.\n- Provides concise behavior guidelines and a clear decision tree to enforce searching live web before answering factual or timely queries.\n- Simplifies command structure for web, news, images, books, and videos search, with improved region and timeframe controls.\n- Details universal integration (OpenClaw, MCP server, LangChain, OpenAI) for LLM agents and pip install instructions.\n- Removes older implementation/agent/script files and replaces existing SKILL.md with comprehensive new documentation focused on agent-centric behavior.\n- Stronger emphasis on real-time accuracy, cite-every-fact policy, and privacy-first design.\n\nv4.0.0 | 2026-03-13T12:06:19.308Z | user\n\nv4.0: Added verify_claim.py for multi-source cross-validation. Now tells you how much to trust the answer with VERIFIED/LIKELY_TRUE/UNCERTAIN/LIKELY_FALSE verdicts. Multi-engine parallel search + 4-tier anti-bot cascade + authority-weighted scoring.\n\nv1.0.0 | 2026-03-13T11:19:36.365Z | user\n\nInitial release - Free, private web search for OpenClaw with self-hosted SearXNG + Scrapling anti-bot fetcher. Zero API keys, zero cost.\n\nArchive index:\n\nArchive v23.0.0: 3 files, 3467 bytes\n\nFiles: skill-card.md (2517b), SKILL.md (3971b), _meta.json (144b)\n\nFile v23.0.0:SKILL.md\n\n---\nname: zero-api-key-web-search\nversion: \"23.0.0\"\ndescription: >\n  OpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking.\n  No API keys required by default; optional providers can be enabled for stronger coverage.\n  v23: multi-engine SERP (7 engines), Web Unlocker for blocked pages, auto-fallback on 403/429.\nhomepage: https://github.com/wd041216-bit/zero-api-key-web-search\n---\n\n# Zero-API-Key Web Search for OpenClaw\n\nThis skill gives OpenClaw a practical verification workflow:\n\n- `zero-search` for live search results (7 engines via Bright Data)\n- `zero-search providers` for provider discovery\n- `zero-browse` for reading pages (auto-unlocks blocked content)\n- `zero-verify` for support/conflict classification\n- `zero-report` for a citation-ready summary with next steps\n- `zero-setup` for interactive provider configuration\n\n## Install\n\n```bash\npip install zero-api-key-web-search\n```\n\n## Minimum verification\n\n```bash\nzero-search \"OpenAI API pricing\" --type news --timelimit w\nzero-search providers\nzero-browse \"https://docs.python.org/3/whatsnew/\"\nzero-verify \"Python 3.13 is the latest stable release\" --deep --max-pages 2 --json\nzero-report \"Python 3.13 stable release\" --claim \"Python 3.13 is the latest stable release\" --deep --json\n```\n\n## Provider paths\n\n| Profile | Providers | Best for |\n| --- | --- | --- |\n| `free` | ddgs | Zero-setup local search |\n| `free-verified` | ddgs, searxng | Free cross-validation |\n| `production` | brightdata | Production reliability and geo-targeting |\n| `production-unlock` | brightdata, web_unlocker | Production SERP + blocked page access |\n| `max-evidence` | ddgs, searxng, brightdata | Maximum provider diversity |\n\n## Recommended flow\n\n1. Run `zero-search` for factual or recent questions.\n2. Use `zero-browse` on the most relevant source when snippets are not enough.\n3. Use `zero-verify` when a concrete claim needs a support/conflict summary.\n4. Use `zero-report` when you want a compact evidence package with citations and next steps.\n5. Use `--deep` when the claim matters enough to justify page-aware verification.\n6. Cite the returned URLs in the final answer.\n7. Use optional `brightdata` only when configured or explicitly requested.\n\n## Multi-engine search (Bright Data)\n\n```bash\nzero-search \"AI regulation\" --provider brightdata --engine google --region us-en --json\nzero-search \"AI regulation\" --provider brightdata --engine bing --region gb-en --json\nzero-search \"news\" --provider brightdata --engine yandex --region ru-ru --json\n```\n\nSupported engines: `google`, `bing`, `duckduckgo`, `yandex`, `baidu`, `yahoo`, `naver`.\n\n## Web Unlocker (blocked pages)\n\n```bash\n# Auto-fallback (default) — direct fetch, then unlocker on 403/429\nzero-browse \"https://protected-site.com/article\"\n\n# Force Web Unlocker\nzero-browse \"https://protected-site.com/article\" --use-unlocker always\n```\n\n## Optional Bright Data provider\n\n```bash\n# Interactive setup wizard\nzero-setup\n\n# Or set environment variables\nexport ZERO_SEARCH_BRIGHTDATA_API_KEY=\"...\"\nexport ZERO_SEARCH_BRIGHTDATA_ZONE=\"serp_api1\"\nexport ZERO_SEARCH_BRIGHTDATA_UNLOCKER_ZONE=\"web_unlocker1\"\n```\n\nNew Bright Data users can sign up at https://get.brightdata.com/h21j9xz4uxgd.\n\n## What success looks like\n\n- the verdict is explicit\n- the result includes support and conflict scores\n- `page_aware` is true when deep verification ran\n- the recommended free path is `ddgs + self-hosted searxng`\n- optional production path is `brightdata + web_unlocker` via `ZERO_SEARCH_BRIGHTDATA_API_KEY`\n- source URLs are ready to cite\n\n## Limits\n\n- `zero-verify` is heuristic and evidence-aware, not a proof engine.\n- The default provider path is `ddgs`.\n- The recommended free upgrade path is self-hosted `searxng` via `ZERO_SEARCH_SEARXNG_URL`.\n- Bright Data is optional and should not receive queries unless configured or requested.\n- Conflicting sources are surfaced, not automatically reconciled.\n\n## License\n\nMIT License.\n\nFile v23.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn7e68xzm01gpx37rcmqtms3gs82t6vt\",\n  \"slug\": \"openclaw-free-web-search\",\n  \"version\": \"23.0.0\",\n  \"publishedAt\": 1777798353463\n}\n\nFile v23.0.0:skill-card.md\n\n## Description:\n\nOpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking, with no API keys required by default and optional providers for stronger coverage.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[wd041216-bit](https://clawhub.ai/user/wd041216-bit)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and agents use this skill to run live web searches, browse source pages, verify concrete claims, and produce citation-ready evidence summaries. It is intended for factual, recent, or source-sensitive work where results should include URLs and support or conflict signals.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The skill asks users to install an unpinned external package whose executable code was not included for review.\n\nMitigation: Install only in an isolated virtual environment, avoid installing as root, and prefer a pinned, hash-verified package version after reviewing the package source.\n\nRisk: Optional Bright Data or other provider credentials may cause providers to receive the relevant queries and URLs.\n\nMitigation: Enable optional providers only when approved for the use case, and provide credentials only when sharing those queries and URLs with the provider is acceptable.\n\nRisk: The documented claim verification is heuristic and may not resolve conflicting sources.\n\nMitigation: Treat verification scores as decision support, review cited sources directly, and preserve surfaced conflicts in downstream answers.\n\n## Reference(s):\n\n- [ClawHub Skill Page](https://clawhub.ai/wd041216-bit/skills/openclaw-free-web-search)\n- [Project Homepage](https://github.com/wd041216-bit/zero-api-key-web-search)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, configuration, guidance]\n\n**Output Format:** [Markdown and JSON-capable command output with source URLs, verification scores, and inline shell commands]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May include support and conflict classifications, page-aware verification status, provider setup guidance, and citation-ready URLs.]\n\n## Skill Version(s):\n\n23.0.0 (source: frontmatter and server release metadata)\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 v21.0.0: 2 files, 1476 bytes\n\nFiles: SKILL.md (2313b), _meta.json (144b)\n\nFile v21.0.0:SKILL.md\n\n---\nname: zero-api-key-web-search\ndescription: >\n  Gemini-ready skill for source-backed web search, page reading, and evidence-aware claim checking.\n  Use it to verify factual answers with live search results and explicit source handling.\nversion: \"21.0.0\"\n---\n\n# Zero-API-Key Web Search\n\nUse this skill when Gemini needs live information, citation-ready sources, or an explicit support/conflict read on a factual claim. The default path is free; optional providers are discoverable for stronger coverage.\n\n## Install\n\nInstall the CLI tools first:\n\n```bash\npip install zero-api-key-web-search\n```\n\n## Core commands\n\n```bash\nzero-search \"latest Python release\" --type news --timelimit w\nzero-search providers\nzero-browse \"https://docs.python.org/3/whatsnew/\"\nzero-verify \"Python 3.13 is the latest stable release\" --deep --max-pages 2 --json\nzero-report \"Python 3.13 stable release\" --claim \"Python 3.13 is the latest stable release\" --deep --json\n```\n\n## When to use\n\n- current facts, releases, versions, and statistics\n- questions that need sources before answering\n- claims that may be supported, contested, or under-evidenced\n- cases where a compact evidence report is more useful than raw search results\n\n## Preferred workflow\n\n1. Start with `zero-search` for current or factual questions.\n2. Use `zero-browse` when snippets are too thin.\n3. Use `zero-verify` for support/conflict classification.\n4. Use `zero-report` when you want one citation-ready artifact.\n5. Prefer the free `ddgs + self-hosted searxng` path for stronger provider diversity.\n6. Use optional `brightdata` only when configured or explicitly requested for production reliability, geo-targeting, or stronger provider diversity.\n\n## Optional Bright Data provider\n\n```bash\nexport ZERO_SEARCH_BRIGHTDATA_API_KEY=\"...\"\nexport ZERO_SEARCH_BRIGHTDATA_ZONE=\"web_search\"  # optional\nzero-search \"AI regulation news\" --provider brightdata --type news --region us-en\n```\n\nNew Bright Data users can sign up at https://get.brightdata.com/h21j9xz4uxgd.\n\n## Limits\n\n- `zero-verify` is heuristic and evidence-aware, not a proof engine.\n- The default provider path is `ddgs`.\n- Bright Data is optional and should not receive queries unless configured or requested.\n- Deep verification is stronger than snippets alone, but still not full-document reasoning.\n\nFile v21.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn7e68xzm01gpx37rcmqtms3gs82t6vt\",\n  \"slug\": \"openclaw-free-web-search\",\n  \"version\": \"21.0.0\",\n  \"publishedAt\": 1777595997068\n}\n\nArchive v18.0.0: 4 files, 8638 bytes\n\nFiles: CHANGELOG.md (8743b), README.md (7053b), SKILL.md (2886b), _meta.json (144b)\n\nFile v18.0.0:SKILL.md\n\n---\nname: zero-api-key-web-search\nversion: \"18.0.0\"\ndescription: >\n  Zero-API-key free web search, browsing, and claim verification for AI agents.\n  No API keys required. 100% free. Designed to reduce hallucination risk by\n  surfacing corroborating and conflicting sources.\nhomepage: https://github.com/wd041216-bit/zero-api-key-web-search\nplatforms:\n  - claude-code\n  - cursor\n  - copilot\n  - gemini\n  - manus\n  - continue\n  - kiro\n  - opencode\n  - codex\n  - openclaw\n  - mcp\n  - cli\n---\n\n# Zero-API-Key Web Search\n\nUse this skill when a task needs live search results, source-backed verification, or full-page reading. No API keys required. 100% free.\n\n## Install\n\n```bash\npip install zero-api-key-web-search\n```\n\nFor Claude Code, the repository also ships `.claude/skills/zero-api-key-web-search/SKILL.md`.\nFor Manus-style Agent Skills workflows, use this root `SKILL.md` plus [docs/manus.md](docs/manus.md).\n\n## Core Commands\n\n```bash\nzero-search \"latest Python release\" --type news --timelimit w\nzero-browse \"https://docs.python.org/3/whatsnew/\"\nzero-verify \"Python 3.13 is the latest stable release\" --json\nzero-report \"Python 3.13 stable release\" --claim \"Python 3.13 is the latest stable release\" --deep --json\n```\n\nLegacy aliases: `zero-search`, `zero-browse`, `zero-verify`, `zero-report`.\n\n## When to Use\n\n- facts, versions, releases, dates, or statistics\n- recent or time-sensitive questions\n- claim checking with citations\n- compact evidence reports with citation-ready source digests\n- tasks where conflicting sources should be surfaced instead of hidden\n- free dual-provider verification with `ddgs + self-hosted searxng`\n\n## Operating Guidance\n\n- Treat `zero-verify` as a first-pass evidence classifier, not a proof engine.\n- Prefer `zero-report` when you need a single artifact that combines verdict, citations, and next steps.\n- Prefer `zero-search --type news` for recent events.\n- Use `zero-browse` when snippets are too thin to justify an answer.\n- Cite URLs for factual claims.\n- If support and conflict are both present, present the disagreement rather than collapsing it.\n\n## Compatibility Names\n\n- Repository: `zero-api-key-web-search`\n- Package: `zero-api-key-web-search`\n- Module: `zero_api_key_web_search`\n- CLI: `zero-search`, `zero-browse`, `zero-verify`, `zero-report`\n- MCP: `zero-mcp`\n- Legacy aliases: `search-web`, `browse-page`, `verify-claim`, `evidence-report`, `cross-validated-search-mcp`, `free-web-search-mcp`\n- Legacy modules: `free_web_search`, `zero_api_key_web_search_compat`\n\n## Limits\n\n- Default provider diversity is limited because the default path uses `ddgs`.\n- The recommended free upgrade path is self-hosted `searxng` via `ZERO_SEARCH_SEARXNG_URL`.\n- Scoring is heuristic and depends on returned snippets.\n- TLS verification is enabled by default; insecure mode requires an explicit environment variable.\n\n## License\n\nMIT License.\n\nFile v18.0.0:README.md\n\n<div align=\"center\">\n  <h1>Zero-API-Key Web Search</h1>\n  <p><strong>Free web search, browsing & claim verification for AI agents.</strong></p>\n  <p><em>No API keys. No accounts. No limits. 100% free.</em></p>\n\n  [![PyPI](https://img.shields.io/pypi/v/zero-api-key-web-search?label=pypi)](https://pypi.org/project/zero-api-key-web-search/)\n  [![Python](https://img.shields.io/pypi/pyversions/zero-api-key-web-search)](https://python.org)\n  [![MCP](https://img.shields.io/badge/MCP-Ready-0f766e.svg)](https://modelcontextprotocol.io/)\n  [![Tests](https://img.shields.io/badge/tests-86%20passing-22c55e.svg)](./tests)\n  [![License](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE)\n</div>\n\n---\n\n## What is this?\n\nA single `pip install` gives your AI agent live web search, full-page reading, and evidence-aware claim verification — all without any API key, account, or paid service.\n\n- **Search**: Live results from DuckDuckGo (free) + optional SearXNG (self-hosted, free)\n- **Browse**: Extract clean text from any URL, stripping boilerplate automatically\n- **Verify**: Classify claims as supported / contested / likely false with evidence scores\n- **Report**: Generate citation-ready evidence reports with rationale and next steps\n\n## Quick start\n\n```bash\npip install zero-api-key-web-search\n\n# Search the web\nzero-search \"Python 3.13 release\" --json\n\n# Read a page\nzero-browse \"https://docs.python.org/3/whatsnew/\" --json\n\n# Verify a claim\nzero-verify \"Python 3.13 is the latest stable release\" --deep --json\n\n# Full evidence report\nzero-report \"Python 3.13 stable release\" \\\n  --claim \"Python 3.13 is the latest stable release\" --deep --json\n```\n\nLegacy CLI aliases (`search-web`, `browse-page`, `verify-claim`, `evidence-report`) also work.\n\n## Why this over a plain search wrapper?\n\n| Feature | Plain search | Zero-API-Key Web Search |\n| --- | --- | --- |\n| Live search results | ✅ | ✅ |\n| News / images / videos / books | ❌ | ✅ |\n| Region & time filtering | ❌ | ✅ |\n| Full-page text extraction | ❌ | ✅ |\n| Claim verification with evidence scores | ❌ | ✅ |\n| Supporting vs. conflicting evidence | ❌ | ✅ |\n| Citation-ready evidence reports | ❌ | ✅ |\n| Dual-provider cross-validation | ❌ | ✅ |\n| API key required | Often | **Never** |\n| Cost | Sometimes | **Free** |\n\n## MCP server\n\nWorks with Claude Code, Cursor, Copilot, and any MCP-compatible agent:\n\n```json\n{\n  \"mcpServers\": {\n    \"zero-api-key-web-search\": {\n      \"command\": \"zero-mcp\"\n    }\n  }\n}\n```\n\nFour tools exposed: `search_web`, `browse_page`, `verify_claim`, `evidence_report`.\n\n## Platform support\n\n| Platform | Status | Entry point |\n| --- | --- | --- |\n| **CLI** | Ready | `zero-search`, `zero-browse`, `zero-verify`, `zero-report` |\n| **MCP** | Ready | `zero-mcp` |\n| **Claude Code** | Ready | `.claude/skills/zero-api-key-web-search/SKILL.md` |\n| **Gemini** | Ready | `GEMINI.md` + `.gemini/SKILL.md` |\n| **Cursor** | Ready | `.cursor/rules/zero-api-key-web-search.md` |\n| **Copilot** | Ready | `.github/copilot/instructions.md` |\n| **Codex** | Ready | `.codex/SKILL.md` |\n| **Continue** | Ready | `.continue/skills/zero-api-key-web-search/SKILL.md` |\n| **Manus** | Ready | Root `SKILL.md` + `docs/manus.md` |\n| **Kiro** | Ready | `.kiro/steering/zero-api-key-web-search.md` |\n| **OpenClaw** | Ready | `zero_api_key_web_search/skills/SKILL.md` |\n\n## How verification works\n\n`zero-verify` uses the **evidence-aware heuristic v3** model:\n\n1. Search for the claim across available providers\n2. Score each source on keyword overlap, source quality, freshness\n3. Classify as supporting, conflicting, or neutral\n4. Optionally fetch top pages for deeper page-aware analysis\n5. Render a verdict with confidence and evidence breakdown\n\n| Verdict | Meaning |\n| --- | --- |\n| `supported` | Strong evidence, low conflict |\n| `likely_supported` | Leans positive, not decisive |\n| `contested` | Support and conflict both meaningful |\n| `likely_false` | Conflict strong, support weak |\n| `insufficient_evidence` | Too weak for any firmer verdict |\n\nThis is a heuristic evidence classifier, not a proof engine. See [docs/trust-model.md](docs/trust-model.md) for details and limitations.\n\n## Free dual-provider setup\n\nDefault install uses DuckDuckGo. For stronger cross-validated evidence, add a free self-hosted SearXNG:\n\n```bash\n./scripts/start-searxng.sh\nexport ZERO_SEARCH_SEARXNG_URL=\"http://127.0.0.1:8080\"\n./scripts/validate-free-path.sh\n```\n\nOr with Docker Compose:\n\n```bash\ncp .env.searxng.example .env\ndocker compose -f docker-compose.searxng.yml up -d\n```\n\nFull guide: [docs/searxng-self-hosted.md](docs/searxng-self-hosted.md).\n\n## Evidence report example\n\n```json\n{\n  \"verdict\": \"contested\",\n  \"confidence\": \"MEDIUM\",\n  \"executive_summary\": \"Evidence is split...\",\n  \"verdict_rationale\": [\"Source A supports...\", \"Source B contradicts...\"],\n  \"coverage_warnings\": [\"Single-provider evidence path.\"],\n  \"source_digest\": [\n    {\"title\": \"...\", \"url\": \"...\", \"classification\": \"supporting\", \"evidence_strength\": 3}\n  ],\n  \"next_steps\": [\"Add a second provider for cross-validation.\"]\n}\n```\n\n## Architecture\n\n```\nzero_api_key_web_search/\n  core.py              # UltimateSearcher — search, verify, report engine\n  browse_page.py       # Readability-style page text extraction\n  mcp_server.py        # MCP server (4 tools)\n  transport.py         # SSL/TLS helpers\n  search_web.py        # CLI: zero-search\n  browse_page.py       # CLI: zero-browse\n  verify_claim.py      # CLI: zero-verify\n  evidence_report.py   # CLI: zero-report\n  providers/\n    base.py            # SearchProvider protocol (sync + async)\n    ddgs.py            # DuckDuckGo provider\n    searxng.py         # SearXNG provider\n  skills/\n    SKILL.md           # Bundled OpenClaw skill\n```\n\nKey engineering features:\n\n- **Circuit breaker**: Trips after 3 consecutive provider failures, auto-resets after 60s\n- **Async support**: `asearch()` for concurrent provider calls via `asyncio.gather`\n- **Structured logging**: Configurable logging at search/verify/report entry points\n- **Readability heuristic**: Semantic HTML5 + ARIA roles + text density scoring\n- **Baseline comparison**: Majority-vote and keyword-count baselines in reports\n- **Sub-claim decomposition**: Targeted sub-queries for independent evidence gathering\n\n## Installation\n\n```bash\npip install zero-api-key-web-search\n```\n\nPython 3.10+ required. No API keys, no accounts, no configuration needed.\n\n## Development\n\n```bash\npip install -e \".[dev]\"\npython -m pytest tests/ -x           # 86 tests\nruff check zero_api_key_web_search/ tests/\npyright zero_api_key_web_search/     # 0 errors\ncoverage report --fail-under=80       # 85% coverage\n```\n\n## Verification for ecosystem reviewers\n\n1. `zero-report \"Python 3.13 stable release\" --claim \"Python 3.13 is the latest stable release\" --deep --json`\n2. [docs/ecosystem-readiness.md](docs/ecosystem-readiness.md)\n3. [docs/gemini-submission-checklist.md](docs/gemini-submission-checklist.md)\n4. [docs/trust-model.md](docs/trust-model.md)\n\n## License\n\nMIT License.\n\nFile v18.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn7e68xzm01gpx37rcmqtms3gs82t6vt\",\n  \"slug\": \"openclaw-free-web-search\",\n  \"version\": \"18.0.0\",\n  \"publishedAt\": 1776636463611\n}\n\nFile v18.0.0:CHANGELOG.md\n\n# Changelog\n\nAll notable changes to this project will be documented in this file.\n\nThe format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),\nand this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).\n\n## [Unreleased]\n\n## [18.0.0] - 2026-04-19\n\n### Changed\n- Renamed the public GitHub repository target to `zero-api-key-web-search`.\n- Unified skill and package metadata at version `18.0.0` for the Clawhub release.\n- Updated Docker, Gemini, and platform skill references to the canonical `zero-*` commands and package name.\n\n### Fixed\n- Replaced remaining visible `Cross-Validated Search` branding in docs, source text, and SVG assets with `Zero-API-Key Web Search`.\n- Fixed the Docker image to copy the active `zero_api_key_web_search` packages and start `zero-mcp`.\n\n## [16.0.0] - 2026-03-24\n\n### Changed\n- Renamed the canonical PyPI package to `zero-api-key-web-search` and the canonical Python module to `zero_api_key_web_search`.\n- Kept `free_web_search` imports and `free-web-search-mcp` available as compatibility aliases for existing integrations.\n- Switched packaging, Docker, CI, and platform docs to point at the canonical v16 names.\n\n### Fixed\n- Aligned README, contributing docs, and bundled skill docs with the current published package and CLI surface.\n- Corrected the repository license wording to match the actual `LICENSE` file.\n- Updated the test suite to validate the active `zero_api_key_web_search.search_web` and `browse_page` implementations.\n- Updated CI to fail on real test regressions instead of continuing after errors.\n- Consolidated search logic into a shared core to reduce drift between `search_web.py` and `search.py`.\n- Switched TLS handling to secure-by-default with explicit opt-in insecure mode for constrained environments.\n- Added CLI and MCP smoke tests plus a modern `pyproject.toml` packaging source of truth.\n- Tightened public docs and platform instructions so they describe the current evidence-aware implementation instead of overstating guarantees.\n\n### Added\n- Added `verify-claim`, a first-pass evidence workflow that classifies search results as supporting, conflicting, or neutral and returns a structured verdict.\n- Added `evidence-report`, a higher-level workflow that combines search, verification, citation-ready source digests, and recommended next steps.\n- Added MCP support for `verify_claim` so agent clients can request claim verification directly.\n- Added MCP support for `evidence_report` so agent clients can request a compact evidence package directly.\n- Added a self-hosted SearXNG setup guide for the recommended free dual-provider path.\n- Added `docker-compose.searxng.yml`, `scripts/start-searxng.sh`, and `scripts/validate-free-path.sh` for lower-friction local SearXNG bootstrapping.\n- Added trust-model, release, support, and security docs plus reproducible quickstart examples.\n- Added build and wheel-install validation to CI for release-grade packaging checks.\n- Added evidence-aware verification metadata such as score breakdown, provider diversity, and contention summaries.\n- Added an explainable evidence model document at `docs/verification-model.md`.\n\n### Changed\n- Upgraded `verify-claim` to `evidence-aware-heuristic-v2` with source-quality, freshness, and domain-diversity scoring.\n- Expanded CLI and MCP outputs to expose weighted support/conflict signals for auditability.\n- Expanded the flagship CLI, MCP, README, and platform skill surfaces to expose `evidence-report` as the preferred reviewer-facing workflow.\n- Improved free-path guidance so warnings and docs now point directly to self-hosted `searxng` as the recommended second provider.\n- Updated README and skill surfaces to describe the current evidence model without overclaiming beyond the implementation.\n\n## [15.0.0] - 2026-03-20\n\n### Added\n- **Universal IDE Support**: Now works with 10+ platforms out of the box\n  - Claude Code: `.claude-plugin/SKILL.md` with UserPromptSubmit hooks\n  - Cursor: `.cursor/rules/zero-api-key-web-search.md`\n  - GitHub Copilot: `.github/copilot/instructions.md`\n  - Gemini CLI: `.gemini/SKILL.md`\n  - Continue: `.continue/skills/zero-api-key-web-search/SKILL.md`\n  - Kiro: `.kiro/steering/zero-api-key-web-search.md`\n  - OpenCode: `.opencode/instructions.md`\n  - Codex: `.codex/SKILL.md`\n  - OpenClaw: `free_web_search/skills/SKILL.md` (existing)\n  - MCP: `free-web-search-mcp` (existing)\n  - CLI: `search-web`, `browse-page` (existing)\n- **Platform Detection Metadata**: Added `platforms` field to `_meta.json`\n- **IDE-Specific Hooks**: Claude Code now has UserPromptSubmit hooks for automatic factual query detection\n- **Comprehensive README**: Updated with all supported platforms and installation instructions\n\n### Changed\n- Version bumped to 15.0.0\n- README completely rewritten to emphasize universal IDE support\n- SKILL.md now lists all supported platforms in frontmatter\n- Package description updated to mention universal IDE support\n\n### Migration Guide\n\n**No breaking changes** — all existing functionality preserved. New IDE configurations are additive.\n\nTo enable IDE-specific features:\n- Claude Code: Install package, `.claude-plugin/SKILL.md` is auto-detected\n- Cursor: Copy `.cursor/rules/` to your project\n- Copilot: `.github/copilot/instructions.md` is auto-detected by VS Code\n- Gemini: Install package, `.gemini/SKILL.md` is auto-detected\n\n## [14.0.0] - 2026-03-19\n\n### Changed\n- Repository branding shifted toward **Zero-API-Key Web Search** and the GitHub repository moved to `wd041216-bit/zero-api-key-web-search`.\n- The published compatibility surface remained stable:\n  - package: `free-web-search-ultimate`\n  - CLI: `search-web`, `browse-page`\n  - MCP server: `free-web-search-mcp`\n  - Python module: `free_web_search`\n\n### Added\n- **Anti-Hallucination Focus**: Repositioned as the primary solution for preventing LLM hallucinations\n- **Confidence Scoring System**: Explicit confidence levels (✅ Verified, 🟢 Likely True, 🟡 Uncertain, 🔴 Likely False)\n- **Cross-Validation Guarantees**: Every fact is verified against multiple independent sources\n- **Conflict Detection**: Automatically flags conflicting information across sources\n- Enhanced documentation with decision tree for agents\n- New keywords: hallucination, cross-validation, verification, anti-hallucination\n\n### Changed\n- Focus shifted from \"free search\" to \"hallucination prevention\"\n- All documentation updated to emphasize cross-validation and anti-hallucination\n- Confidence scoring now explicit: HIGH (3+ sources), MEDIUM (2 sources), LOW (1 source)\n- Repository description updated to highlight anti-hallucination capabilities\n\n### Migration Guide\n\n**No install or command migration was required in v14.x.**\n\nThe repository branding changed, but the supported install and command surface stayed:\n\n```bash\npip install free-web-search-ultimate\nsearch-web \"query\"\n```\n\n## [13.0.0] - 2026-03-18\n\n### Changed\n- Fully internationalized all user-facing strings, error messages, and code comments to English for universal compatibility\n- Rewrote all module, class, and function docstrings to Google style with complete `Args:`, `Returns:`, and `Example:` sections\n- Expanded all multi-parameter function signatures to multi-line format per PEP 8 and Google style guide\n- `UltimateSearcher.search()` docstring now documents all `timelimit` values (`d/w/m/y`) and `search_type` options\n- `_cross_validate()` now uses typed `Dict[str, List[Source]]` annotation\n- Error fallback message changed to English: \"No results found. The search engine may be rate-limited.\"\n- CLI output formatting improved: unified spacing, cleaner REPL prompt\n- `browse_page.py` module docstring updated; all inline comments translated to English\n- Version bumped to 13.0.0\n\n## [12.0.0] - 2026-03-18\n\n### Added\n- Published to PyPI: `pip install free-web-search-ultimate`\n- SKILL.md updated to use PyPI installation (eliminates supply-chain risk from git-based install)\n- Python 3.10+ requirement enforced in CI and package metadata\n\n### Changed\n- Installation method changed from `pip install git+...` to `pip install free-web-search-ultimate`\n\n## [11.0.0] - 2025-03-17\n\n### Added\n- Universal Search-First Knowledge Acquisition Plugin for LLMs\n- Full MCP (Model Context Protocol) server support\n- CLI-Anything standard compliance with PATH-accessible entry points\n- REPL interactive mode for continuous search sessions\n- `SKILL.md` auto-discovery bundled with Python package\n\n### Changed\n- Removed suspicious language from `SKILL.md`, replaced with guideline-based approach\n- Improved security posture: no hardcoded credentials, no third-party routing\n\n### Fixed\n- Removed dead Yahoo engine\n- Fixed thread safety issues in concurrent search\n\nArchive v16.0.0: 2 files, 1324 bytes\n\nFiles: SKILL.md (2008b), _meta.json (144b)\n\nFile v16.0.0:SKILL.md\n\n---\nname: cross-validated-search\nversion: \"16.0.0\"\ndescription: >\n  OpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking.\n  Use it to verify factual answers with live search results and explicit source handling.\nhomepage: https://github.com/wd041216-bit/cross-validated-search\n---\n\n# Cross-Validated Search for OpenClaw\n\nThis skill gives OpenClaw a practical verification workflow:\n\n- `search-web` for live search results\n- `browse-page` for reading the full content of a source\n- `verify-claim` for support/conflict classification\n- `evidence-report` for a citation-ready summary with next steps\n\n## Install\n\n```bash\npip install cross-validated-search\n```\n\n## Minimum verification\n\n```bash\nsearch-web \"OpenAI API pricing\" --type news --timelimit w\nverify-claim \"Python 3.13 is the latest stable release\" --deep --max-pages 2 --json\nevidence-report \"Python 3.13 stable release\" --claim \"Python 3.13 is the latest stable release\" --deep --json\n```\n\n## Recommended flow\n\n1. Run `search-web` for factual or recent questions.\n2. Use `browse-page` on the most relevant source when snippets are not enough.\n3. Use `verify-claim` when a concrete claim needs a support/conflict summary.\n4. Use `evidence-report` when you want a compact evidence package with citations and next steps.\n5. Use `--deep` when the claim matters enough to justify page-aware verification.\n6. Cite the returned URLs in the final answer.\n\n## What success looks like\n\n- the verdict is explicit\n- the result includes support and conflict scores\n- `page_aware` is true when deep verification ran\n- the recommended free path is `ddgs + self-hosted searxng`\n- source URLs are ready to cite\n\n## Limits\n\n- `verify-claim` is heuristic and evidence-aware, not a proof engine.\n- The default provider path is `ddgs`.\n- The recommended free upgrade path is self-hosted `searxng` via `CROSS_VALIDATED_SEARCH_SEARXNG_URL`.\n- Conflicting sources are surfaced, not automatically reconciled.\n\n## License\n\nMIT License.\n\nFile v16.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn7e68xzm01gpx37rcmqtms3gs82t6vt\",\n  \"slug\": \"openclaw-free-web-search\",\n  \"version\": \"16.0.0\",\n  \"publishedAt\": 1774318077011\n}\n\nArchive v11.0.0: 2 files, 2279 bytes\n\nFiles: SKILL.md (4004b), _meta.json (144b)\n\nFile v11.0.0:SKILL.md\n\n---\nname: free-web-search-ultimate\nversion: \"11.0.0\"\ndescription: >\n  Universal Search-First Knowledge Acquisition Plugin for LLMs.\n  Enables real-time web search and deep page browsing via MCP or CLI.\n  Zero-cost, privacy-first, supports text/news/images/books/videos.\nhomepage: https://github.com/wd041216-bit/free-web-search-ultimate\n---\n\n# Free Web Search Ultimate v11.0 — Search-First Paradigm\n\n## Overview\n\nThis skill enables real-time web search and page browsing for any LLM agent. It introduces a **Search-First** approach to knowledge acquisition: instead of relying solely on training data, the agent should use live web search as the primary source for factual, technical, and time-sensitive information.\n\n## Core Behavior Guidelines\n\nWhen this skill is active, the agent should follow these guidelines:\n\n**Guideline 1 — Search Before Answering**: For questions involving facts, current events, software versions, statistics, people, companies, or technical documentation, call `search-web` before generating a response.\n\n**Guideline 2 — Treat Training Data as a Starting Point**: Training data has a cutoff date. For any claim that may have changed recently, verify it via search.\n\n**Guideline 3 — Deep-Dive When Needed**: If search snippets don't provide enough detail, use `browse-page` to read the full source document.\n\n**Guideline 4 — Always Cite Sources**: Every factual statement in the final response should be accompanied by the source URL from search results.\n\n**Guideline 5 — Prefer News for Recency**: For anything that happened in the last year, use `search-web --type news` to get the most recent information.\n\n## Available Commands\n\n### `search-web` — Real-Time Web Search\n\n```bash\n# General knowledge (default)\nsearch-web \"query here\"\n\n# Current events and news\nsearch-web \"query here\" --type news\n\n# Images\nsearch-web \"query here\" --type images\n\n# Academic / books\nsearch-web \"query here\" --type books\n\n# Videos\nsearch-web \"query here\" --type videos\n\n# Region-specific (e.g., Chinese)\nsearch-web \"查询内容\" --region zh-cn\n\n# Time-limited (d=day, w=week, m=month, y=year)\nsearch-web \"query here\" --timelimit w\n\n# Machine-readable JSON output\nsearch-web \"query here\" --json\n```\n\n### `browse-page` — Deep Page Reading\n\n```bash\n# Read full content of a URL\nbrowse-page \"https://example.com/article\"\n\n# JSON output\nbrowse-page \"https://example.com/article\" --json\n```\n\n## Decision Tree for Agents\n\n```\nUser asks a question\n        │\n        ▼\nIs it purely creative/hypothetical?\n   YES → Answer directly\n   NO  ▼\nDoes it involve facts, events, versions, or data?\n   YES ▼\nRun: search-web \"<query>\" [--type news if recent event]\n        │\n        ▼\nAre snippets sufficient to answer?\n   YES → Synthesize answer + cite sources\n   NO  ▼\nRun: browse-page \"<top_result_url>\"\n        │\n        ▼\nSynthesize answer from full page content + cite source\n```\n\n## Why Search-First?\n\n| Default LLM Behavior | Search-First Behavior |\n|---|---|\n| Answers from training data | Answers from live web |\n| Knowledge cutoff applies | Always up-to-date |\n| May produce outdated facts | Cites verifiable sources |\n| Single knowledge source | Multi-source cross-validation |\n\n## Integration\n\nThis skill works with:\n- **OpenClaw** — native skill integration\n- **Claude Desktop / Cursor** — via MCP server (`free-web-search-mcp`)\n- **LangChain** — via Python tool wrappers\n- **OpenAI Function Calling** — via JSON schema definitions\n\n## MCP Server Setup (Claude Desktop / Cursor)\n\nAdd to your `claude_desktop_config.json`:\n\n```json\n{\n  \"mcpServers\": {\n    \"free-web-search\": {\n      \"command\": \"free-web-search-mcp\",\n      \"args\": []\n    }\n  }\n}\n```\n\nInstall via pip from the GitHub repository:\n\n```bash\npip install git+https://github.com/wd041216-bit/free-web-search-ultimate.git\n```\n\n## Requirements\n\n- Python 3.8+\n- `beautifulsoup4`, `lxml`, `ddgs`, `mcp>=1.1.2`\n\n## License\n\nMIT-0 — Free to use, modify, and redistribute. No attribution required.\n\nFile v11.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn7e68xzm01gpx37rcmqtms3gs82t6vt\",\n  \"slug\": \"openclaw-free-web-search\",\n  \"version\": \"11.0.0\",\n  \"publishedAt\": 1773774017644\n}\n\nArchive v4.0.0: 5 files, 16322 bytes\n\nFiles: AGENTS.md (1997b), scripts/browse_page.py (18292b), scripts/search_local_web.py (19277b), SKILL.md (5124b), _meta.json (143b)\n\nFile v4.0.0:SKILL.md\n\n---\nname: local-web-search\ndescription: >\n  Free, private, real-time web search for OpenClaw — zero API keys required.\n  Powered by self-hosted SearXNG + Scrapling anti-bot engine. Multi-engine\n  parallel search (Bing/DuckDuckGo/Google/Startpage/Qwant), intent-aware\n  Agent Reach query expansion, three-tier Browse/Viewing (Fetcher →\n  StealthyFetcher → DynamicFetcher for Cloudflare/JS sites), cross-engine\n  anti-hallucination validation, and automatic public fallback.\nhomepage: https://github.com/wd041216-bit/openclaw-free-web-search\nmetadata:\n  clawdbot:\n    emoji: \"🔍\"\n    requires:\n      env: []\n    files: [\"scripts/*\"]\n---\n\n# Local Free Web Search v3.0\n\nUse this skill when the user needs current or real-time web information.\nPowered by **Scrapling** (anti-bot) + **SearXNG** (self-hosted search).\nZero API keys. Zero cost. Runs entirely locally.\n\n---\n\n## External Endpoints\n\n| Endpoint | Data Sent | Purpose |\n|---|---|---|\n| `http://127.0.0.1:18080` (local) | Search query string only | Local SearXNG instance |\n| `https://searx.be` (fallback only) | Search query string only | Public fallback when local SearXNG is down |\n| Any URL passed to `browse_page.py` | HTTP GET request only | Fetch page content for reading |\n\nNo personal data, no credentials, no conversation history is ever sent to any endpoint.\n\n---\n\n## Security & Privacy\n\n- All search queries go to your **local SearXNG** instance by default — no third-party tracking\n- Public fallback (`searx.be`) is only used when local service is unavailable, and only receives the raw query string\n- `browse_page.py` makes standard HTTP GET requests to URLs you explicitly pass — no data is posted\n- Scrapling runs entirely locally — no cloud API calls, no telemetry\n- No API keys required or stored\n- No conversation history or personal data leaves your machine\n\n**Trust Statement:** This skill sends search queries to your local SearXNG instance (default) or `searx.be` (fallback). Page content is fetched via standard HTTP GET. No personal data is transmitted. Only install if you trust the public SearXNG instance at `searx.be` as a fallback.\n\n---\n\n## Model Invocation Note\n\nThis skill is invoked autonomously by the agent when a query requires live web information. You can disable autonomous invocation by removing this skill from your workspace. The agent will only use this skill when it determines real-time information is needed.\n\n---\n\n## Tool 1 — Web Search\n\n```bash\npython3 ~/.openclaw/workspace/skills/local-web-search/scripts/search_local_web.py \\\n  --query \"YOUR QUERY\" \\\n  --intent general \\\n  --limit 5\n```\n\n**Intent options** (controls engine selection + query expansion):\n\n| Intent | Best for |\n|---|---|\n| `general` | Default, mixed queries |\n| `factual` | Facts, definitions, official docs |\n| `news` | Latest events, breaking news |\n| `research` | Papers, GitHub, technical depth |\n| `tutorial` | How-to guides, code examples |\n| `comparison` | A vs B, pros/cons |\n| `privacy` | Sensitive queries (ddg/startpage/qwant only) |\n\n**Additional flags:**\n\n| Flag | Description |\n|---|---|\n| `--engines bing,duckduckgo,...` | Override engine selection |\n| `--freshness hour\\|day\\|week\\|month\\|year` | Filter by recency |\n| `--max-age-days N` | Downrank results older than N days |\n| `--browse` | Auto-fetch top result with browse_page.py |\n| `--no-expand` | Disable Agent Reach query expansion |\n| `--json` | Machine-readable JSON output |\n\n---\n\n## Tool 2 — Browse/Viewing (read full page)\n\n```bash\npython3 ~/.openclaw/workspace/skills/local-web-search/scripts/browse_page.py \\\n  --url \"https://example.com/article\" \\\n  --max-words 600\n```\n\n**Fetcher modes** (use `--mode` flag):\n\n| Mode | Fetcher | Use case |\n|---|---|---|\n| `auto` | Tier 1 → 2 → 3 | Default — tries fast first |\n| `fast` | `Fetcher` | Normal sites |\n| `stealth` | `StealthyFetcher` | Cloudflare / anti-bot sites |\n| `dynamic` | `DynamicFetcher` | Heavy JS / SPA sites |\n\nReturns: title, published date, word count, confidence (HIGH/MEDIUM/LOW),\nfull extracted text, and anti-hallucination advisory.\n\n---\n\n## Recommended Workflow\n\n1. Run `search_local_web.py` — review results by Score and `[cross-validated]` tag\n2. Run `browse_page.py` on the top URL — check Confidence level\n3. If Confidence is LOW (paywall/blocked) — retry with `--mode stealth` or try next URL\n4. Answer only after reading HIGH-confidence page content\n5. **Never state facts from snippets alone**\n\n---\n\n## Rules\n\n- Always use `--intent` to match the query type for best results.\n- When local SearXNG is unavailable, both scripts automatically fall back to `searx.be`.\n- If the fallback also fails, tell the user to start local SearXNG:\n\n```bash\ncd \"$(cat ~/.openclaw/workspace/skills/local-web-search/.project_root)\" && ./start_local_search.sh\n```\n\n- Do NOT invent search results if all sources fail.\n- `search_local_web.py` and `browse_page.py` are complementary: **search first, browse second**.\n- Prefer `[cross-validated]` results (appeared in multiple engines) for factual claims.\n- For sites behind Cloudflare or requiring JS, use `browse_page.py --mode stealth`.\n\nFile v4.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn7e68xzm01gpx37rcmqtms3gs82t6vt\",\n  \"slug\": \"openclaw-free-web-search\",\n  \"version\": \"4.0.0\",\n  \"publishedAt\": 1773403579308\n}\n\nFile v4.0.0:AGENTS.md\n\n# Agent Injection Rules — Local Free Web Search v2.0\n\n## When to use this skill\n\nActivate this skill when:\n- The user asks about current events, latest news, recent releases, or real-time information\n- The built-in `web_search` tool is unavailable or not configured\n- The user explicitly asks to search the web\n- The task requires verifying facts with up-to-date sources\n\n## Step 1 — Search\n\nRun the search script with the appropriate intent:\n\n```bash\npython3 ~/.openclaw/workspace/skills/local-web-search/scripts/search_local_web.py \\\n  --query \"<user query>\" \\\n  --intent <factual|news|research|tutorial|comparison|privacy|general> \\\n  --limit 5\n```\n\nChoose intent based on query type:\n- Current events / news → `--intent news --freshness day`\n- Technical docs / facts → `--intent factual`\n- Academic / GitHub → `--intent research`\n- How-to / examples → `--intent tutorial`\n- A vs B questions → `--intent comparison`\n- Sensitive queries → `--intent privacy`\n\n## Step 2 — Browse top results\n\nFor each result with Score > 50 or marked `[cross-validated]`:\n\n```bash\npython3 ~/.openclaw/workspace/skills/local-web-search/scripts/browse_page.py \\\n  --url \"<result URL>\" \\\n  --max-words 600\n```\n\n## Step 3 — Answer rules\n\n- **HIGH confidence** → safe to answer from this page\n- **MEDIUM confidence** → answer with caveat, suggest checking another source\n- **LOW confidence** → do NOT use this page; try next URL\n- **NEVER** fabricate content when all sources fail — tell the user the search returned no usable results\n- **ALWAYS** cite the URL and published date when answering factual questions\n- **PREFER** `[cross-validated]` results over single-engine results for factual claims\n\n## Fallback behaviour\n\nIf local SearXNG is unavailable, both scripts automatically fall back to `searx.be`.\nIf all sources fail, output:\n\n```\nSearch unavailable. To start local SearXNG:\ncd \"$(cat ~/.openclaw/workspace/skills/local-web-search/.project_root)\" && ./start_local_search.sh\n```\n\nArchive v1.0.0: 5 files, 16321 bytes\n\nFiles: AGENTS.md (1997b), scripts/browse_page.py (18292b), scripts/search_local_web.py (19277b), SKILL.md (5124b), _meta.json (143b)\n\nFile v1.0.0:SKILL.md\n\n---\nname: local-web-search\ndescription: >\n  Free, private, real-time web search for OpenClaw — zero API keys required.\n  Powered by self-hosted SearXNG + Scrapling anti-bot engine. Multi-engine\n  parallel search (Bing/DuckDuckGo/Google/Startpage/Qwant), intent-aware\n  Agent Reach query expansion, three-tier Browse/Viewing (Fetcher →\n  StealthyFetcher → DynamicFetcher for Cloudflare/JS sites), cross-engine\n  anti-hallucination validation, and automatic public fallback.\nhomepage: https://github.com/wd041216-bit/openclaw-free-web-search\nmetadata:\n  clawdbot:\n    emoji: \"🔍\"\n    requires:\n      env: []\n    files: [\"scripts/*\"]\n---\n\n# Local Free Web Search v3.0\n\nUse this skill when the user needs current or real-time web information.\nPowered by **Scrapling** (anti-bot) + **SearXNG** (self-hosted search).\nZero API keys. Zero cost. Runs entirely locally.\n\n---\n\n## External Endpoints\n\n| Endpoint | Data Sent | Purpose |\n|---|---|---|\n| `http://127.0.0.1:18080` (local) | Search query string only | Local SearXNG instance |\n| `https://searx.be` (fallback only) | Search query string only | Public fallback when local SearXNG is down |\n| Any URL passed to `browse_page.py` | HTTP GET request only | Fetch page content for reading |\n\nNo personal data, no credentials, no conversation history is ever sent to any endpoint.\n\n---\n\n## Security & Privacy\n\n- All search queries go to your **local SearXNG** instance by default — no third-party tracking\n- Public fallback (`searx.be`) is only used when local service is unavailable, and only receives the raw query string\n- `browse_page.py` makes standard HTTP GET requests to URLs you explicitly pass — no data is posted\n- Scrapling runs entirely locally — no cloud API calls, no telemetry\n- No API keys required or stored\n- No conversation history or personal data leaves your machine\n\n**Trust Statement:** This skill sends search queries to your local SearXNG instance (default) or `searx.be` (fallback). Page content is fetched via standard HTTP GET. No personal data is transmitted. Only install if you trust the public SearXNG instance at `searx.be` as a fallback.\n\n---\n\n## Model Invocation Note\n\nThis skill is invoked autonomously by the agent when a query requires live web information. You can disable autonomous invocation by removing this skill from your workspace. The agent will only use this skill when it determines real-time information is needed.\n\n---\n\n## Tool 1 — Web Search\n\n```bash\npython3 ~/.openclaw/workspace/skills/local-web-search/scripts/search_local_web.py \\\n  --query \"YOUR QUERY\" \\\n  --intent general \\\n  --limit 5\n```\n\n**Intent options** (controls engine selection + query expansion):\n\n| Intent | Best for |\n|---|---|\n| `general` | Default, mixed queries |\n| `factual` | Facts, definitions, official docs |\n| `news` | Latest events, breaking news |\n| `research` | Papers, GitHub, technical depth |\n| `tutorial` | How-to guides, code examples |\n| `comparison` | A vs B, pros/cons |\n| `privacy` | Sensitive queries (ddg/startpage/qwant only) |\n\n**Additional flags:**\n\n| Flag | Description |\n|---|---|\n| `--engines bing,duckduckgo,...` | Override engine selection |\n| `--freshness hour\\|day\\|week\\|month\\|year` | Filter by recency |\n| `--max-age-days N` | Downrank results older than N days |\n| `--browse` | Auto-fetch top result with browse_page.py |\n| `--no-expand` | Disable Agent Reach query expansion |\n| `--json` | Machine-readable JSON output |\n\n---\n\n## Tool 2 — Browse/Viewing (read full page)\n\n```bash\npython3 ~/.openclaw/workspace/skills/local-web-search/scripts/browse_page.py \\\n  --url \"https://example.com/article\" \\\n  --max-words 600\n```\n\n**Fetcher modes** (use `--mode` flag):\n\n| Mode | Fetcher | Use case |\n|---|---|---|\n| `auto` | Tier 1 → 2 → 3 | Default — tries fast first |\n| `fast` | `Fetcher` | Normal sites |\n| `stealth` | `StealthyFetcher` | Cloudflare / anti-bot sites |\n| `dynamic` | `DynamicFetcher` | Heavy JS / SPA sites |\n\nReturns: title, published date, word count, confidence (HIGH/MEDIUM/LOW),\nfull extracted text, and anti-hallucination advisory.\n\n---\n\n## Recommended Workflow\n\n1. Run `search_local_web.py` — review results by Score and `[cross-validated]` tag\n2. Run `browse_page.py` on the top URL — check Confidence level\n3. If Confidence is LOW (paywall/blocked) — retry with `--mode stealth` or try next URL\n4. Answer only after reading HIGH-confidence page content\n5. **Never state facts from snippets alone**\n\n---\n\n## Rules\n\n- Always use `--intent` to match the query type for best results.\n- When local SearXNG is unavailable, both scripts automatically fall back to `searx.be`.\n- If the fallback also fails, tell the user to start local SearXNG:\n\n```bash\ncd \"$(cat ~/.openclaw/workspace/skills/local-web-search/.project_root)\" && ./start_local_search.sh\n```\n\n- Do NOT invent search results if all sources fail.\n- `search_local_web.py` and `browse_page.py` are complementary: **search first, browse second**.\n- Prefer `[cross-validated]` results (appeared in multiple engines) for factual claims.\n- For sites behind Cloudflare or requiring JS, use `browse_page.py --mode stealth`.\n\nFile v1.0.0:_meta.json\n\n{\n  \"ownerId\": \"kn7e68xzm01gpx37rcmqtms3gs82t6vt\",\n  \"slug\": \"openclaw-free-web-search\",\n  \"version\": \"1.0.0\",\n  \"publishedAt\": 1773400776365\n}\n\nFile v1.0.0:AGENTS.md\n\n# Agent Injection Rules — Local Free Web Search v2.0\n\n## When to use this skill\n\nActivate this skill when:\n- The user asks about current events, latest news, recent releases, or real-time information\n- The built-in `web_search` tool is unavailable or not configured\n- The user explicitly asks to search the web\n- The task requires verifying facts with up-to-date sources\n\n## Step 1 — Search\n\nRun the search script with the appropriate intent:\n\n```bash\npython3 ~/.openclaw/workspace/skills/local-web-search/scripts/search_local_web.py \\\n  --query \"<user query>\" \\\n  --intent <factual|news|research|tutorial|comparison|privacy|general> \\\n  --limit 5\n```\n\nChoose intent based on query type:\n- Current events / news → `--intent news --freshness day`\n- Technical docs / facts → `--intent factual`\n- Academic / GitHub → `--intent research`\n- How-to / examples → `--intent tutorial`\n- A vs B questions → `--intent comparison`\n- Sensitive queries → `--intent privacy`\n\n## Step 2 — Browse top results\n\nFor each result with Score > 50 or marked `[cross-validated]`:\n\n```bash\npython3 ~/.openclaw/workspace/skills/local-web-search/scripts/browse_page.py \\\n  --url \"<result URL>\" \\\n  --max-words 600\n```\n\n## Step 3 — Answer rules\n\n- **HIGH confidence** → safe to answer from this page\n- **MEDIUM confidence** → answer with caveat, suggest checking another source\n- **LOW confidence** → do NOT use this page; try next URL\n- **NEVER** fabricate content when all sources fail — tell the user the search returned no usable results\n- **ALWAYS** cite the URL and published date when answering factual questions\n- **PREFER** `[cross-validated]` results over single-engine results for factual claims\n\n## Fallback behaviour\n\nIf local SearXNG is unavailable, both scripts automatically fall back to `searx.be`.\nIf all sources fail, output:\n\n```\nSearch unavailable. To start local SearXNG:\ncd \"$(cat ~/.openclaw/workspace/skills/local-web-search/.project_root)\" && ./start_local_search.sh\n```","readmeExcerpt":"Skill: Zero Api Key Web Search Owner: wd041216-bit Summary: OpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking. No API keys required by default; optional providers can be enab... Tags: fact-check:21.0.0, latest:23.0.0, mcp:21.0.0, no-api-key:21.0.0, openclaw:21.0.0, v11:11.0.0, verification:21.0.0, web-search:21.0.0 Version history: v23.0.0 | 2026-05-03T08:52:33.463Z | user v","codeSnippets":[],"executableExamples":[{"language":"bash","snippet":"pip install zero-api-key-web-search"},{"language":"bash","snippet":"zero-search \"OpenAI API pricing\" --type news --timelimit w\nzero-search providers\nzero-browse \"https://docs.python.org/3/whatsnew/\"\nzero-verify \"Python 3.13 is the latest stable release\" --deep --max-pages 2 --json\nzero-report \"Python 3.13 stable release\" --claim \"Python 3.13 is the latest stable release\" --deep --json"},{"language":"bash","snippet":"zero-search \"AI regulation\" --provider brightdata --engine google --region us-en --json\nzero-search \"AI regulation\" --provider brightdata --engine bing --region gb-en --json\nzero-search \"news\" --provider brightdata --engine yandex --region ru-ru --json"},{"language":"bash","snippet":"# Auto-fallback (default) — direct fetch, then unlocker on 403/429\nzero-browse \"https://protected-site.com/article\"\n\n# Force Web Unlocker\nzero-browse \"https://protected-site.com/article\" --use-unlocker always"},{"language":"bash","snippet":"# Interactive setup wizard\nzero-setup\n\n# Or set environment variables\nexport ZERO_SEARCH_BRIGHTDATA_API_KEY=\"...\"\nexport ZERO_SEARCH_BRIGHTDATA_ZONE=\"serp_api1\"\nexport ZERO_SEARCH_BRIGHTDATA_UNLOCKER_ZONE=\"web_unlocker1\""},{"language":"bash","snippet":"pip install zero-api-key-web-search"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: zero-api-key-web-search\nversion: \"23.0.0\"\ndescription: >\n  OpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking.\n  No API keys required by default; optional providers can be enabled for stronger coverage.\n  v23: multi-engine SERP (7 engines), Web Unlocker for blocked pages, auto-fallback on 403/429.\nhomepage: https://github.com/wd041216-bit/zero-api-key-web-search\n---\n\n# Zero-API-Key Web Search for OpenClaw\n\nThis skill gives OpenClaw a practical verification workflow:\n\n- `zero-search` for live search results (7 engines via Bright Data)\n- `zero-search providers` for provider discovery\n- `zero-browse` for reading pages (auto-unlocks blocked content)\n- `zero-verify` for support/conflict classification\n- `zero-report` for a citation-ready summary with next steps\n- `zero-setup` for interactive provider configuration\n\n## Install\n\n```bash\npip install zero-api-key-web-search\n```\n\n## Minimum verification\n\n```bash\nzero-search \"OpenAI API pricing\" --type news --timelimit w\nzero-search providers\nzero-browse \"https://docs.python.org/3/whatsnew/\"\nzero-verify \"Python 3.13 is the latest stable release\" --deep --max-pages 2 --json\nzero-report \"Python 3.13 stable release\" --claim \"Python 3.13 is the latest stable release\" --deep --json\n```\n\n## Provider paths\n\n| Profile | Providers | Best for |\n| --- | --- | --- |\n| `free` | ddgs | Zero-setup local search |\n| `free-verified` | ddgs, searxng | Free cross-validation |\n| `production` | brightdata | Production reliability and geo-targeting |\n| `production-unlock` | brightdata, web_unlocker | Production SERP + blocked page access |\n| `max-evidence` | ddgs, searxng, brightdata | Maximum provider diversity |\n\n## Recommended flow\n\n1. Run `zero-search` for factual or recent questions.\n2. Use `zero-browse` on the most relevant source when snippets are not enough.\n3. Use `zero-verify` when a concrete claim needs a support/conflict summary.\n4. Use `zero-report` when you want a compact evidence package with citations and next steps.\n5. Use `--deep` when the claim matters enough to justify page-aware verification.\n6. Cite the returned URLs in the final answer.\n7. Use optional `brightdata` only when configured or explicitly requested.\n\n## Multi-engine search (Bright Data)\n\n```bash\nzero-search \"AI regulation\" --provider brightdata --engine google --region us-en --json\nzero-search \"AI regulation\" --provider brightdata --engine bing --region gb-en --json\nzero-search \"news\" --provider brightdata --engine yandex --region ru-ru --json\n```\n\nSupported engines: `google`, `bing`, `duckduckgo`, `yandex`, `baidu`, `yahoo`, `naver`.\n\n## Web Unlocker (blocked pages)\n\n```bash\n# Auto-fallback (default) — direct fetch, then unlocker on 403/429\nzero-browse \"https://protected-site.com/article\"\n\n# Force Web Unlocker\nzero-browse \"https://protected-site.com/article\" --use-unlocker always\n```\n\n## Optional Bright Data provider\n\n```bash\n# Interactive setup wizard\nzero-setup\n\n# Or set environment variables"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn7e68xzm01gpx37rcmqtms3gs82t6vt\",\n  \"slug\": \"openclaw-free-web-search\",\n  \"version\": \"23.0.0\",\n  \"publishedAt\": 1777798353463\n}"},{"path":"skill-card.md","content":"## Description:\n\nOpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking, with no API keys required by default and optional providers for stronger coverage.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[wd041216-bit](https://clawhub.ai/user/wd041216-bit)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nDevelopers and agents use this skill to run live web searches, browse source pages, verify concrete claims, and produce citation-ready evidence summaries. It is intended for factual, recent, or source-sensitive work where results should include URLs and support or conflict signals.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The skill asks users to install an unpinned external package whose executable code was not included for review.\n\nMitigation: Install only in an isolated virtual environment, avoid installing as root, and prefer a pinned, hash-verified package version after reviewing the package source.\n\nRisk: Optional Bright Data or other provider credentials may cause providers to receive the relevant queries and URLs.\n\nMitigation: Enable optional providers only when approved for the use case, and provide credentials only when sharing those queries and URLs with the provider is acceptable.\n\nRisk: The documented claim verification is heuristic and may not resolve conflicting sources.\n\nMitigation: Treat verification scores as decision support, review cited sources directly, and preserve surfaced conflicts in downstream answers.\n\n## Reference(s):\n\n- [ClawHub Skill Page](https://clawhub.ai/wd041216-bit/skills/openclaw-free-web-search)\n- [Project Homepage](https://github.com/wd041216-bit/zero-api-key-web-search)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, shell commands, configuration, guidance]\n\n**Output Format:** [Markdown and JSON-capable command output with source URLs, verification scores, and inline shell commands]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May include support and conflict classifications, page-aware verification status, provider setup guidance, and citation-ready URLs.]\n\n## Skill Version(s):\n\n23.0.0 (source: frontmatter and server release metadata)\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."}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"OpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking. No API keys required by default; optional providers can be enab... Skill: Zero Api Key Web Search Owner: wd041216-bit Summary: OpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking. No API keys required by default; optional providers can be enab... Tags: fact-check:21.0.0, latest:23.0.0, mcp:21.0.0, no-api-key:21.0.0, openclaw:21.0.0, v11:11.0.0, verification:21.0.0, web-search:21.0.0 Version history: v23.0.0 | 2026-05-03T08:52:33.463Z | user v","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":1384,"uniquenessScore":48,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-09T07:56:24.950Z","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":"2026-10-09T07:56:24.950Z","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":"2026-10-10T08:09:53.634Z","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"}]}}}