Parkour Mcp
Connects Claude to Kagi, Semantic Scholar, arXiv, GitHub, Wikipedia, Reddit, and other sources through their native APIs rather than scraping. Fetches URLs intelligently by detecting the source and routing to structured endpoints. Returns content as Markdown with YAML frontmatter that steers follow-up requests, like pointing from an arXiv abstract to the HTML full text or Semantic Scholar for citations. Maintains a rolling cache so you can extract sections or search without refetching. Passively tracks citations from papers and repos as you browse. Reach for this when you need Claude to pull research or technical content without hitting rate limits, login walls, or burning tool calls on dead ends.
- GitHub stars
- 4
- tools
- 10
- transport
- STDIO
- last updated
- Aug 21, 2026
Tools
10 toolsPublic tool metadata for what this MCP can expose to an agent.
kagi_searchSearch the web using Kagi.com's curated, SEO-resistant index
Search the web using Kagi.com's curated, SEO-resistant index
No parameter schema in public metadata yet.
web_fetch_sectionsList section headings and anchor slugs for a web page (for targeted extraction). Long documents paginate via `slice=` in 100-section windows
List section headings and anchor slugs for a web page (for targeted extraction). Long documents paginate via `slice=` in 100-section windows
No parameter schema in public metadata yet.
web_fetch_incisiveFetch a Markdown rendered version of a HTML webpage, a PDF, or a markdown file, with section, search, and slice support (JSON, XML, YAML, and plain text are returned raw). `requires_js=true` renders JavaScript-dependent pages through a headless browser, with an `actions` ReAct...
Fetch a Markdown rendered version of a HTML webpage, a PDF, or a markdown file, with section, search, and slice support (JSON, XML, YAML, and plain text are returned raw). `requires_js=true` renders JavaScript-dependent pages through a headless browser, with an `actions` ReAct...
No parameter schema in public metadata yet.
semantic_scholarSearch and retrieve academic paper data from Semantic Scholar (search, paper details, references, authors, body text snippets)
Search and retrieve academic paper data from Semantic Scholar (search, paper details, references, authors, body text snippets)
No parameter schema in public metadata yet.
arxivSearch and retrieve academic papers from arXiv (search with field-prefix syntax, paper details, category browsing)
Search and retrieve academic papers from arXiv (search with field-prefix syntax, paper details, category browsing)
No parameter schema in public metadata yet.
githubSearch and retrieve code, issues, pull requests, commits, and comparisons from GitHub (9 actions: search_issues, search_code, search_repos, repo, tree, issue, pull_request, file, issue_templates)
Search and retrieve code, issues, pull requests, commits, and comparisons from GitHub (9 actions: search_issues, search_code, search_repos, repo, tree, issue, pull_request, file, issue_templates)
No parameter schema in public metadata yet.
ietfSearch and retrieve IETF RFCs and Internet-Drafts (4 actions: rfc, search, draft, subseries)
Search and retrieve IETF RFCs and Internet-Drafts (4 actions: rfc, search, draft, subseries)
No parameter schema in public metadata yet.
packagesInspect software packages across 7 language ecosystems via deps.dev (5 actions: package, version, dependencies, project, advisory)
Inspect software packages across 7 language ecosystems via deps.dev (5 actions: package, version, dependencies, project, advisory)
No parameter schema in public metadata yet.
discourseSearch and browse Discourse forum topics (3 actions: topic, search, latest) — auto-detected via response headers
Search and browse Discourse forum topics (3 actions: topic, search, latest) — auto-detected via response headers
No parameter schema in public metadata yet.
mediawikiSearch and retrieve Wikipedia / MediaWiki articles, with native footnote and inline-citation resolution (3 actions: page, search, references). First tool to use the split `title=` / `query=` parameter convention
Search and retrieve Wikipedia / MediaWiki articles, with native footnote and inline-citation resolution (3 actions: page, search, references). First tool to use the split `title=` / `query=` parameter convention
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
Parkour MCP
parkour
"an activity in which people move quickly around buildings and objects in a city while performing jumps and other skilful movements, usually trying to move between points as quickly, smoothly, and safely as possible"
Parkour is a content exploration toolkit that helps LLMs surface high signal, unsummarized web content. It makes extensive use of clean APIs and Markdown conversion to enable targeted content extraction and knowledge synthesis. A rolling 2Q page cache keeps recently visited pages in memory so that follow-up requests (section extraction, BM25 search, slice retrieval, comparison pivots) are served quickly and without additional round-trips. While primarily designed for Claude Code and Claude Desktop, it should be adaptable to most agentic toolchain needs.
Details
- Repository
- github.com/blightbow/parkour-mcp
- Package
https://github.com/blightbow/parkour-mcp/releases/download/v2.4.0/parkour-mcp.mcpb- Transport
- STDIO
- Registry
- MCP registry, active