> ## Documentation Index > Fetch the complete documentation index at: https://docs.ishlabs.io/llms.txt > Use this file to discover all available pages before exploring further. # docs tools > MCP tool reference for the docs domain. ## `docs_get` **Tier:** read-only · **Tags:** compat, docs, read Read the Ish docs - index of pages or one page body. Returns the same content as the `ish://docs` and `ish://docs/<slug>` MCP resources. Use this if your harness doesn't surface MCP resources (e.g. most spawned-agent contexts in Claude Code as of 2026-05). With no arg, returns the index of available pages. With `slug`, returns the markdown body of that one page. `slug`: optional doc slug, forward-slashed (e.g. `"overview"`, `"overview-mcp"`, `"concepts/study"`, `"reference/aliases"`). Omit to list every slug. **Returns** when `slug` is omitted: `{"pages": [{"slug", "title", "description"}, ...], "by_category": {"overview": [...], "concepts": [...], "guides": [...], "reference": [...]}}` - the flat list plus the same entries bucketed by namespace prefix so agents can scan one category at a time. When `slug` is set: `{"slug": str, "body": str}` where `body` is the page's markdown. Errors: \[validation\_error] for path-traversal-shaped slugs; \[not\_found] when the slug doesn't match any page (the message lists available slugs). ### Parameters