| name | agentics-content |
| description | Retrieve, search, and reason over content on agentics.org via the agentics-org MCP server. Use when a question references agentics.org, the Agentics community, membership, events, or projects, or when content from the public site is needed as ground truth. |
| allowed-tools | mcp__agentics-org__fetch_page mcp__agentics-org__search_content mcp__agentics-org__list_pages |
Agentics Content
Read and search the public agentics.org website through the agentics-org MCP server.
When to use
- The user asks about agentics.org pages, content, navigation, or copy
- Questions about Agentics community, membership tiers, events, projects, or contact info
- A task needs current site content as the source of truth (e.g., "what does the about page say?")
Available MCP tools
fetch_page(path, format?) — fetch a single page; returns stripped text by default
search_content(query, paths?, context_chars?) — search across pages for a substring; returns hits with surrounding context
list_pages() — list known routes (sitemap.xml if available, else curated)
Available resources
agentics://<slug> URIs for home, about, community, events, blog, contact, membership, projects.
Available prompts
summarize-section — summarize a page for a given audience
answer-from-site — answer a question using agentics.org content as ground truth
Workflow
- If the user's request is open-ended ("what's on agentics.org?"), call
list_pages first
- For specific topic searches, call
search_content with the topic; it returns matching paths
- For full content, call
fetch_page on the relevant path
- Cite the URL in your response so the user can verify
Caveats
agentics.org is a SPA — JS-rendered content is not visible to plain HTTP fetch. The MCP server returns the SSR shell + meta tags + any pre-rendered text. If a page comes back nearly empty, that content is rendered client-side; tell the user and suggest opening the page directly.