{
  "markdown": "# wikipedia\n\nA Claude Code plugin for fetching content from Wikipedia.\n\n## Why this exists\n\nClaude Code's built-in `WebFetch` tool returns HTTP 403 on all Wikipedia URLs. Wikipedia's server blocks requests from Python HTTP libraries (`python-requests`, `urllib`) based on their default user-agent strings. This is separate from robots.txt — it is an active server-side filter. WebFetch uses one of these libraries internally and has no parameter to override the user-agent.\n\nThis plugin teaches Claude to use `curl` via the Bash tool instead. Curl's default user-agent (`curl/X.X`) is not blocked.\n\n## Related open issues\n\nIssues on `anthropics/claude-code`:\n\n- [#7568](https://github.com/anthropics/claude-code/issues/7568) — WebFetch uses a generic user-agent (root cause)\n- [#22846](https://github.com/anthropics/claude-code/issues/22846) — WebFetch returns 403 on Wikipedia and other sites\n- [#29009](https://github.com/anthropics/claude-code/issues/29009) — Claude Desktop share URLs return 403\n- [#33985](https://github.com/anthropics/claude-code/issues/33985) — WebFetch: allow access to public claude.ai/share links\n\n## Skills\n\n### `/wikipedia:fetch`\n\nFetches Wikipedia articles using curl. Supports three URL patterns:\n\n- Raw wikitext — wiki markup, good for extracting tables\n- MediaWiki API — structured JSON with plain-text article content\n- REST API — short JSON summary\n\n## Installation\n\n```\n/plugin install wikipedia\n```\n\nOr add from this repository:\n\n```\n/plugin marketplace add andyrosa/claude-code-wikipedia-plugin\n/plugin install wikipedia@andyrosa/claude-code-wikipedia-plugin\n```\n",
  "bytes": 1601,
  "sha": "c29ccc5cc50c2c03a502752dee4cfb52f4092554ab3be19c010ee0b60a61a277",
  "repo_slug": "andyrosa/claude-code-wikipedia-plugin",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_andyrosa_claude_code_wikipedia_plugin_wi_88c40f92/readme"
}