MCP server for the Omni REST API (stdio).
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"omni-app-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# omni-app-mcp [](https://github.com/trustxai/omni-mcp/actions/workflows/ci.yml) [](https://pypi.org/project/omni-app-mcp/) [](LICENSE) An MCP server that puts the **whole Omni REST API** in front of an LLM client over stdio — users, groups and permissions, connections and dbt, models and model YAML, git branches and pull requests, content validation, documents (v1 and v2), folders and labels, queries returning decoded Apache Arrow results, dashboards, schedules, AI jobs, routines, evals and credit governance — as typed, self-documenting tools that return markdown or JSON. Every tool maps to one documented API operation, carries a docstring written for a model choosing what to call, and never raises. ## Why this exists Omni ships [its own MCP server](https://docs.omni.co/ai/mcp/index), and it is good at what it is built for: a remote, OAuth-authenticated, query-focused server that turns natural language into scoped queries against a model and topic. It is deliberately narrow — analysis, not administration. This project covers the other half. It runs locally over stdio against the REST API, so an agent can **administer, model, validate and publish**: branch a model, edit its YAML with checksum conflict detection, validate it, open the pull request, merge the branch, sync git; find every document still pointing at a renamed field and fix them; provision users, groups and model roles; wire dbt; publish document drafts; run and pause schedules; audit AI credit limits. The two are complementary and can be installed side by side — see [docs/comparison.md](docs/comparison.md) for an honest side-by-side. ## Quick start Once published on PyPI: ```bash uvx omni-app-mcp ``` Until then, run it straight from the repository: ```bash uvx --from git+https://github.com/tru…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.