
Odile Labs MCP server
Which MCP servers are Odile Labs's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.
read 2026-10-24
- Registry namecom.odilelabs/gateverified in the MCP registry
The registry name sits under odilelabs.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.
- Stateanswerschecked 2026-10-24
- Tools21
- Server calls itselfcom.odilelabs/gate
- Server version1.6.0
- Protocol version2025-06-18
On 2026-10-24 the server answered initialize and tools/list without a sign-in. First answer in 75 ms.
- Its tools do none of the product's use cases we checked: answer calls and text owners each caller's name, number and need; forward missed calls to a receptionist after hours; reply to customers by text after calls come in.
- Checked on 4 of the 7 days it takes to judge reliability.
Once connected
- gate_create_project, gate_add_app, gate_create_run, gate_set_app_runs, gate_set_response change data without a sign-in: anyone with the address can call them.
Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.
- INITE MCP scoreafter a weeka score needs 7 daily checks: 4 so far
- What an agent can do with it0 of 3 use cases
- Answeredmeasuring4 of the 7 daily checks it takes to print a share
- Median first answer128 ms
- Space in an agent's context≈ 4,451 tokensthe definitions of all its tools (21), which an agent carries on every turn
- Tool definitionsA · 97 of 100
- Typed results0 of 21 toolsdeclared outputSchema: results come typed, read once
- Parameters without a description: in 4 of 21
What an agent can do with it
- cannot · answer calls and text owners each caller's name, number and need
- cannot · forward missed calls to a receptionist after hours
- cannot · reply to customers by text after calls come in
Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?
Worth a look
- Tools that change data, on a server that answers without a sign-in: gate_create_project, gate_add_app, gate_create_run, gate_set_app_runs, gate_set_response
A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →
gate_openshell_setupRead-only local setup plan for native OpenShell + Gate: one command, exact installer hash, recipe schema, prerequisites, authority boundaries and cleanup. Default creates an isolated Free read-only demo with no model or payment. A trusted…
gate_create_projectCreate a Free project in front of an existing HTTPS API. No signup or payment required. For a credential-free demo use https://sandbox.gate.odilelabs.com with GET /customers/:cid/tickets/:id. Returns project_id, admin_key (shown once) and…
gate_add_appGive one app its own key and the calls it may make. Routes are rules like "GET /orders/:id", "POST /orders", "GET /users/{user}/orders" (per-user), or example calls like "GET /orders/42", which become rules allowing any id. Returns key…
gate_list_appsThe project's apps with their allowed calls, limits, state and anomaly-guard settings.
gate_check_setupRead actual Gate protection settings before giving an integration production access. Reports observe-only mode, writes without mandatory runs, missing user-token configuration, wildcard grants and approval separation, with evidence, exact…
gate_metricsRead last-24-hour stored call counts by HTTP status and reason, measured duration through response headers (including upstream wait), per-app anomaly scores and 24-hour anomaly counts, and separate current-month gateway usage: plan, used…
gate_auditThe most recent calls through the gateway (time, app, matched route, status, reason, duration), newest first, with totals by status and refusal reason. Gate never stores bodies, query values or headers.
gate_create_runGate Actions: give one agent session narrow, expiring permission. Actions are rules with {placeholders} bound by scope (e.g. "PATCH /customers/{customer}/tickets/:id" with scope {"customer":"A"}), each with an optional max and approval…
gate_set_app_runsMake an app need a run: "writes" (every write needs x-odile-run), "all" (every call), or "off".
gate_set_responseOpt-in JSON response redaction: return only explicit JSON Pointer paths, e.g. /id and /items/*/name. * matches one segment; selecting a container allows its whole subtree. Applies to every upstream response for this app, including errors…
gate_add_approverCreate an approver key for the person who approves held actions (shown once). Give it to that person, never to an agent: once a project has an approver, the admin key can no longer approve.
gate_list_actionsActions waiting for approval (or in another state), with the exact call. With an installed approval-risk model, pending actions sort by risk descending and include risk (0–1) and risk_features (top two advisory signals). No model means no…
gate_explain_refusalWhat a Gate refusal means and how to fix it, from the x-gate-refused header value (e.g. route_not_allowed).
gate_delegate_runHand a narrower run to another active app in the same project. Uses the parent app key and run token, never the admin key. Scope, expiry, approvals and ancestor budgets stay enforced; maximum depth five.
gate_learn_suggestionsRead normalized routes and a suggested per-minute limit from an admin-enabled, time-boxed learning session. Records only method, normalized path templates and counts per minute; no query values, raw ids, headers or bodies. Never enables…
gate_audit_proofRead chained audit rows, decided-action feature snapshots and anchors with an HMAC-SHA256 signature. No sealed paths/bodies, notes, keys or IPs in the action labels. Legacy actions lack snapshots and are excluded. store_audit=false has no…
gate_audit_verifyRecompute a pasted export chain and project signature; returns intact or broken_at (one-based row position). Does not approve actions or change permissions.
gate_run_treeRead a run and its delegated descendants without tokens.
gate_run_timelineRead call metadata, approvals and approver names, recovery outcomes and delegated children; no bodies or query values.
gate_validate_configParse and check a declarative Gate config (apps, routes, rate limits, run mode, plugins including anomaly-guard, alerts, approvers) without changing anything. anomaly-guard settings: mode off|alert|strict, z_threshold 2–10, min_minutes…
gate_sync_declarativeMake a Gate project match an odile.yaml: creates the project (if no project_id) and missing apps and approvers, applies run mode, loop, learning and anomaly-guard settings and alerts. Example plugin: {name: anomaly-guard, settings: {mode…
How to connect
Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:
{
"mcpServers": {
"gate": {
"type": "http",
"url": "https://gate.odilelabs.com/mcp"
}
}
}Add it
Claude Code, in a terminal
claude mcp add --transport http odilelabs https://gate.odilelabs.com/mcpclaude.ai, ChatGPT and other clients: add https://gate.odilelabs.com/mcp as a custom connector in the settings.
- com.odilelabs/odileanswers, 23 toolsverified in the MCP registry
OpenAPI spec
12 paths, read 2026-10-24. Open the spec ↗
Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.
Cookies
Before we count this visit
Analytics and ad measurement tell us which pages and which of our ads do their job: Google Analytics and Google Ads, the Meta pixel, Yandex Metrica, and one first-party id of our own. Nothing is stored until you answer, and declining changes nothing about the site.