generated: '2026-08-29' method: probed source: >- https://www.checklyhq.com/.well-known/mcp.json (200), https://www.checklyhq.com/.well-known/mcp/server-card.json (200), https://api.checklyhq.com/.well-known/oauth-protected-resource (200), https://www.checklyhq.com/docs/ai/mcp-server (200) and its tools reference https://www.checklyhq.com/docs/ai/mcp-server/tools (200). A live POST of tools/list to https://api.checklyhq.com/mcp returned HTTP 401 with an RFC 9728 WWW-Authenticate challenge, so the tool list below is transcribed from Checkly's published tools reference, not from an authenticated introspection. name: Checkly MCP Server description: >- Hosted, remote MCP server that lets an agent inspect Checkly checks, check results and test sessions, fetch result assets, read and trigger Rocky AI root cause analyses, manage account environment variables, and read status pages and manage their incidents. deployment: mode: remote endpoint: https://api.checklyhq.com/mcp transport: streamable-http auth: oauth verified: probed note: >- OAuth 2.1 is the documented and recommended path, restricted to Checkly-approved MCP clients; user and service API keys are supported for clients that can send custom headers. There is no stdio package - repeated npm probes for @checkly/mcp-server, @checkly/mcp and checkly-mcp all returned 404. Agents that need to author or deploy check code use the Checkly CLI instead; the docs state the MCP server runs remotely and cannot reach a local filesystem. manifests: server_manifest: https://www.checklyhq.com/.well-known/mcp.json server_card: https://www.checklyhq.com/.well-known/mcp/server-card.json documentation: https://www.checklyhq.com/docs/ai/mcp-server/ agent_skills_index: https://www.checklyhq.com/.well-known/agent-skills/index.json authentication: schemes: - oauth2 - bearer protected_resource_metadata: https://api.checklyhq.com/.well-known/oauth-protected-resource authorization_servers: - https://auth.checklyhq.com/ bearer_methods_supported: - header account_selection_header: X-Checkly-Account scopes: - checkly:account:read - checkly:account:invite - checkly:checks:read - checkly:checks:write - checkly:checks:run - checkly:incidents:read - checkly:incidents:write - checkly:environment-variables:read - checkly:environment-variables:write - checkly:status-pages:read - checkly:rca:read - checkly:rca:run - checkly:test-sessions:read - checkly:assets:read live_probe: url: https://api.checklyhq.com/mcp method: POST body: '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' http_status: 401 www_authenticate: >- Bearer error="invalid_token", error_description="Authentication required", resource_metadata="https://api.checklyhq.com/.well-known/oauth-protected-resource/mcp" note: >- A textbook RFC 9728 challenge - the 401 names the protected-resource document and the full required scope set, so an MCP client can complete discovery unauthenticated. tools: - name: whoami category: account type: read description: Return the authenticated user and account memberships visible to the MCP session. - name: get-account-entitlements category: account type: read description: Return the resolved account plan and feature entitlement state. - name: list-account-members category: account type: read description: List account members and pending invites. - name: invite-account-member category: account type: write description: >- Invite a user to the resolved account. OAuth-only, requires Owner or Admin access, sends an invite email, and is explicitly documented as not idempotent. - name: prepare-local-check-authoring category: account type: read description: Return an account-aware runbook for creating, testing and deploying check code with the local Checkly CLI. - name: list-check-stats category: checks type: read description: >- Paginated, filterable list of checks with pass/fail/degraded status; includeReliability adds availability, response-time percentiles and ICMP latency or packet loss. - name: list-check-results category: checks type: read description: Recent compact results for one check, including response status and assertion counts. - name: get-check-result category: checks type: read description: Compact detail for one check result. - name: get-check-performance-summary category: checks type: read description: Average, P50, P95 and P99 response-time metrics for one check over a date range. - name: trigger-checks category: checks type: write description: Run existing deployed checks on demand and record the run as a test session. Consumes check-run execution quota. - name: list-test-sessions category: test-sessions type: read description: List recent test sessions with filters and cursor ordering. - name: get-test-session category: test-sessions type: read description: Return one test session and compact result rows. - name: list-test-session-results category: test-sessions type: read description: List compact result rows for one test session with bounded pagination. - name: get-test-session-result category: test-sessions type: read description: Compact detail for one test-session result. - name: list-check-result-assets category: assets type: read description: Normalized asset manifest for a check result. - name: fetch-check-result-asset category: assets type: read description: Fetch one check-result asset; text inline under a byte limit, binary or large assets as a download link. - name: list-test-session-result-assets category: assets type: read description: Normalized asset manifest for a test-session result. - name: fetch-test-session-result-asset category: assets type: read description: Fetch one test-session-result asset. - name: get-error-group-root-cause-analyses category: rca type: read description: Return a check or test-session error group and any existing root cause analyses. - name: get-root-cause-analysis category: rca type: read description: Return one RCA by ID; pending analyses return compact polling status. - name: trigger-root-cause-analysis category: rca type: write description: Trigger a new RCA for an error group. Requires run access and consumes RCA invocation quota. - name: list-account-environment-variables category: environment-variables type: read description: List account-level variables and secrets; secret values return as null. - name: get-account-environment-variable category: environment-variables type: read description: Get one account-level variable by key; secret values return as null. - name: create-account-environment-variable category: environment-variables type: write description: Create an account-level variable or secret; secret values are encrypted and never echoed back. - name: update-account-environment-variable category: environment-variables type: write description: Update an account-level variable or secret by key. - name: list-status-pages category: status-pages type: read description: List status pages including the cards and services needed for incident targeting. - name: get-status-page category: status-pages type: read description: Get one status page with cards and services. - name: list-status-page-incidents category: incidents type: read description: List status page incidents, optionally filtered by page and status. - name: get-status-page-incident category: incidents type: read description: Get one incident with services and incident updates. - name: create-status-page-incident category: incidents type: write description: Create a new incident and optionally notify subscribers. Not idempotent. - name: update-status-page-incident category: incidents type: write description: Post a progress update to an existing incident and optionally notify subscribers. Not idempotent. - name: resolve-status-page-incident category: incidents type: write description: Resolve an incident by posting a final resolved update and optionally notifying subscribers. Not idempotent. tool_count: 32 counts: read: 24 write: 8 notes: - >- Tools are surfaced conditionally - the docs state each tool appears only when the MCP session holds the matching permission, and some write tools additionally require an Admin or Read & Write account role. A tool list from an authenticated session will therefore be a subset of this one. - >- Checkly documents the division of labour explicitly - MCP for live account state and actions, Agent Skills plus the CLI for authoring, testing and deploying check code.