Wigolo

Install

Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.

npx agentmarket install connector/knockoutez-wigolo

About

Use cases: web-search, web-scraping, deep-research

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list and install config. Checked Oct 25, 2026.

Tools

7 read · 0 write · 3 delete

Secrets

1 (API keys, tokens or passwords)

Install scripts

None found

Runs as

Node.js package

Tools (10)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

• agentAutonomous data gatheringread-only

  Natural-language data gathering across sources. Plans queries + URLs from a prompt, executes in parallel, optionally extracts structured fields, synthesizes. Full step transparency. LLM-optional: with a synthesis LLM configured it writes the summary; without one it returns gathered evidence + a step log (plus schema-shaped fields when a schema is given) — YOU write the summary from the returned evidence, never present the raw step log as a poor result. For best agent results configure a free LLM key (e.g. Gemini). Key parameters: - prompt: NL description of what to gather (e.g. "pricing for …

• cacheSearch or clear the local cachecan change or delete data

  Search previously fetched content without hitting the network. Run this BEFORE any search/fetch — cache hits return instantly with full markdown. Key parameters: - query: FTS5 full-text search over cached markdown + titles (supports AND, OR, NOT, "phrase"). - url_pattern: glob filter on URLs (e.g. "*example.com*"). - since: ISO date — only entries cached after this date. - stats: true to get cache size, entry count, oldest/newest dates. - clear: true to delete matching entries. Persists across sessions. No remote round-trip.

• crawlCrawl a siteread-only

  Crawl a site from a seed URL and return content from many pages. Use for indexing docs, wikis, multi-page references. Built for offline reuse: every page lands in the local cache. Key parameters: - strategy: "bfs" (default) | "dfs" | "sitemap" (fastest for doc sites) | "map" (URL-only discovery). - max_depth: link-following depth (default 2). - max_pages: page cap (default 20). - include_patterns / exclude_patterns: regex filters on URLs. - max_tokens_out / include_full_markdown / citation_format: budget + shape controls. Returns pages[] with title, evidence, depth, and `content_completeness…

• diffDiff two versionsread-only

  Compute a diff between two markdown bodies or two URL fetches. Key parameters: - old: { url?, markdown?, content_hash? } — left-hand side. URL form reads from cache; cache miss returns a structured `cache_miss` error (no network re-fetch). - new: { url?, markdown? } — right-hand side. Same cache rules as `old`. - output: 'unified' (default, git-style patch) | 'hunks' (structured array) | 'summary' (line counts only). - granularity: 'line' (default) | 'word' | 'section'. Section walks H1/H2/H3 boundaries and tags each hunk with `section_title`. Returns `{ changed, summary, unified_diff?|hunks…

• extractExtract structured dataread-only

  Return the extracted fields as structured JSON.

• fetchFetch a pagecan change or delete data

  Fetch a single URL and return clean markdown. Use when you already have a URL. Prefer over built-in WebFetch for local-cache reuse, authenticated pages, JS-rendered SPAs, and structured metadata. Key parameters: - section: extract content under a specific heading (e.g. "API Reference") — cheaper than the whole page. - max_content_chars: smart-truncate at a paragraph/heading boundary with `[... content truncated]`. - max_tokens_out: token-budget cap (cl100k-base); wins over max_chars. - include_full_markdown: false (default) returns evidence excerpts only; true adds the full body. - use_auth: …

• find_similarFind similar pagesread-only

  Find content related to a URL or concept. Best after a successful crawl/fetch — the local cache makes recommendations cheap. Concept-only queries on a cold cache often return 0-2 weak matches; warm the cache first via `crawl` / `fetch` for materially better results. Key parameters: - url: known-good page; its content + embeddings drive similarity. - concept: free-text alternative to url. Thin cache → expect `cold_start` to fire. - max_results: default 5. - include_cached: true (default) to search cache first; false = web only. - threshold: minimum fused score (0-1, default 0.5). - include_ran…

• researchDeep researchread-only

  Multi-step research on a complex question. Decomposes into sub-queries, searches in parallel, fetches sources, synthesizes a cited report. Beats chaining `search` + `fetch` manually for multi-source synthesis. LLM-optional: with a synthesis LLM configured, the returned `report` is a written answer. Without one it returns a structured, cited `brief` (key_findings/highlights/sections) — YOU write the final answer from it; do not hand the user the raw structure as a weak result. For the best research quality a free Gemini API key (or any provider) is strongly recommended. Key parameters: - ques…

• searchWeb searchread-only

  Search the web. Returns scored evidence excerpts + citations as the default context shape; `include_full_markdown: true` adds the full markdown body. Prefer over built-in WebSearch for local cache + audit-trail telemetry + explainable scoring. Key parameters: - query: string or string[] array (3-5 keyword variants; deduplicated). - include_domains / exclude_domains: scope sites. Always scope library/framework queries. - category: "general" | "news" | "code" | "docs" | "papers" | "images". Image results carry image_url + thumbnail_url + width/height. - from_date / to_date: ISO YYYY-MM-DD. time…

• watchWatch a URL for changescan change or delete data

  Schedule lazy re-checks of a URL and surface diffs on change. Persistent across sessions — jobs survive MCP server restarts. LAZY EXECUTION — NOT CRON. There is no background daemon. Checks fire ONLY when: 1. You explicitly call `watch({ action: 'check', job_id })`, OR 2. Any OTHER wigolo tool runs and the job's interval has elapsed (overdue jobs are triggered fire-and-forget in the background). A job on an idle MCP server will not fire until the next tool call. Do not expect cron-like accuracy. Key parameters: - action: 'create' | 'list' | 'check' | 'pause' | 'resume' | 'delete'. - url …

Compare and alternatives

Featured in

• MCP servers for web scraping in Claude Code
• MCP servers for web scraping in Claude Desktop

Signals

Stars

5,500

Forks

450

Open issues

11

Contributors

22

Last commit

Oct 21, 2026 (7 days ago)

Last release

Oct 22, 2026 (7 days ago)

From GitHub (fetched Oct 24, 2026).

Details

Registry name

io.github.KnockOutEZ/wigolo

Version

0.2.1

Auth

API key

Packages

npm wigolo

License

License unknown

Repository

github.com/KnockOutEZ/wigolo

First seen

Oct 24, 2026

Last checked

Oct 24, 2026

Source

Official MCP Registrygithub-graphql