CNCF release intelligence: typed facts from release notes; check_stack compares versions locally.
Ratatosk MCP server exists for a simple reason — assistants are far more useful when they can act on Ratatosk directly instead of describing what you should do. CNCF release intelligence: typed facts from release notes; check_stack compares versions locally.
In Norse myth, Ratatoskr is the squirrel that carries messages up and down the world tree. This one carries release intelligence. ratatosk.io watches 76 CNCF projects and turns every release note into typed, entity-level facts: security fixes, breaking changes, removals, deprecations, changed defaults. Plain bug fixes and marketing copy are filtered out. What remains is what an operator acts on.
Once Ratatosk is connected, these are the calls the assistant has available:
check_stack — Takes the component versions you run and returns the facts on your upgrade path — critical/high split from the rest, one entry per advisory, eachlist_facts — The incremental fact feed, oldest-analyzed first. Filter by project, type, or severity; page with the since cursor to keep a local copy in syncfacts_by_entity — Reverse lookup: every fact touching one exact identifier — such as a CVE id, CRD, feature gate, flag, config field, or dependencyget_release — One release in full: its facts, an overall assessment, and the link to the original note. A fully reviewed release with zero facts means it was readlist_releases — The newest releases of one project as one-line summaries (date, fact counts by severity), newest first — the tool for "what changed in X lately"list_projects — The roster of tracked projects and their canonical slugs (the short project id every other tool takes) — look names up here instead of guessingBeing a remote server, there is no local install. You register the endpoint with your client, authorise it once, and the tools appear.
Plenty of planning and project tracking servers cover similar ground. The differences that matter in practice are scope of access and how much setup stands between you and a working tool call. Ratatosk's toolset — check_stack, list_facts, facts_by_entity and 3 more — is a fair guide to whether it matches your workflow. It is maintained by garlicKim21; worth a glance at recent repository activity before you build anything load-bearing on it.
We check each listing at SyncDev against the project's documentation before it goes live — if something here drifts out of date, it is a bug worth reporting.
| Tool | What it does |
|---|---|
| check_stack | Takes the component versions you run and returns the facts on your upgrade path — critical/high split from the rest, one entry per advisory, each with its quote and ids. The comparison happens inside the server process — |
| list_facts | The incremental fact feed, oldest-analyzed first. Filter by project, type, or severity; page with the since cursor to keep a local copy in sync |
| facts_by_entity | Reverse lookup: every fact touching one exact identifier — such as a CVE id, CRD, feature gate, flag, config field, or dependency |
| get_release | One release in full: its facts, an overall assessment, and the link to the original note. A fully reviewed release with zero facts means it was read and found routine |
| list_releases | The newest releases of one project as one-line summaries (date, fact counts by severity), newest first — the tool for "what changed in X lately" |
| list_projects | The roster of tracked projects and their canonical slugs (the short project id every other tool takes) — look names up here instead of guessing |
{
"mcpServers": {
"ratatosk": {
"command": "docker",
"args": ["run", "-i", "--rm", "ghcr.io/garlickim21/ratatosk-mcp"]
}
}
}Add to claude_desktop_config.json, then restart Claude Desktop.
Your workspace, on speaking terms with AI — search, read and write Notion pages and databases.
Issue tracking at the speed of conversation — Linear's official hosted server with OAuth and zero install.
Search and read your Drive — Docs, Sheets and files become context your assistant can actually use.
Your assistant inside the workspace — read channels, search history, post messages and tame the noise.
Metabase ships its own MCP endpoint — search your BI content, build and run queries, and save questions and dashboards without leaving the chat.
Stop letting your assistant hallucinate n8n node parameters — this server hands it the real schemas, templates and validation.