generated: '2026-08-14'
method: probed
status: published
source: https://docs.exec.com/mcp
summary: 'Exec ships a real, anonymous, remote MCP server on its own documentation host: https://docs.exec.com/mcp.
  It answers initialize/tools/list/resources/list over Streamable HTTP (text/event-stream) with no credentials.
  It is a DOCUMENTATION server, not a wrapper over the Exec REST API — its three tools search and read
  docs.exec.com and file docs feedback; none of them call api.exec.com/rest/v1. The REST surface has no
  MCP server of its own, so the one-tool-per-operation list derived from the OpenAPI is retained separately
  below as candidates.'
deployment:
  mode: none
  verified: derived
  tools: 32
  checked: '2026-08-12'
  source: catalog MCP census
server:
  name: Exec Docs
  version: 1.0.0
  transport: http
  protocol_version: '2025-11-25'
  url: https://docs.exec.com/mcp
  capabilities:
    tools:
      listChanged: true
    resources:
      listChanged: true
  instructions_excerpt: '"This Model Context Protocol server provides search and retrieval tools for the
    Exec Docs site... Apart from the submit_feedback tool, the server is read-only and scoped to Exec
    Docs."'
tools:
- name: search_exec_docs
  title: Search documentation
  description: Search across the Exec Docs knowledge base to find relevant information, code examples,
    API references, and guides.
  input_schema:
    type: object
    required:
    - query
    properties:
      query:
        type: string
        description: Search query
  annotations:
    readOnlyHint: true
    destructiveHint: false
    idempotentHint: true
    openWorldHint: false
- name: query_docs_filesystem_exec_docs
  description: Run a read-only shell-like query (rg/grep/find/tree/ls/cat/head/tail/jq/...) against a
    virtualized in-memory filesystem containing only the Exec Docs pages and OpenAPI specs. Stateless
    per call; output truncated to 30KB.
  input_schema:
    type: object
    required:
    - command
    properties:
      command:
        type: string
        description: A shell command to run against the virtualized documentation filesystem
  annotations:
    readOnlyHint: true
    destructiveHint: false
    idempotentHint: true
    openWorldHint: false
- name: submit_feedback
  title: Submit documentation feedback
  description: Report a problem with a documentation page so the docs team can fix it.
  input_schema:
    type: object
    required:
    - path
    - feedback
    properties:
      path:
        type: string
        minLength: 1
        description: The documentation page path the feedback is about
      feedback:
        type: string
        minLength: 1
        description: A clear description of the documentation issue or suggestion
  annotations:
    readOnlyHint: false
resources:
- uri: mintlify://skills/exec
  name: exec
  mime_type: text/markdown
  description: Provider-published Agent Skill playbook for Exec (name/description frontmatter + product
    summary, API quick reference, decision guidance, gotchas, verification checklist). Saved verbatim
    to skills/exec-provider-skill.md.
candidate_rest_tools:
  status: candidate
  method: derived
  source: openapi/_original/exec-openapi-original.yml
  note: No MCP server exposes the Exec REST API. These are one-tool-per-operation candidates derived from
    the OpenAPI, not tools anyone can call today.
  tools:
  - name: getWorkspace
    description: Get workspace info
  - name: listWorkspaceMembers
    description: List workspace members
  - name: listWorkspaceGroups
    description: List workspace groups
  - name: listAssignments
    description: List scenario assignments
  - name: listCollections
    description: List collections
  - name: listSessions
    description: List roleplay sessions
  - name: getSession
    description: Get session detail
  - name: listSkills
    description: List skills
  - name: getSkillProficiency
    description: Get skill proficiency by user
  - name: getScenarioAnalyticsSummary
    description: Get scenario analytics summary
  - name: getScenarioAnalyticsParticipants
    description: Get scenario participant analytics
  - name: listScenarios
    description: List scenarios
  - name: checkScenarioAccess
    description: Check scenario access
  - name: grantScenarioAccess
    description: Grant scenario access
  - name: assignScenario
    description: Assign scenario to user
  - name: createScenarioStudioSession
    description: Create Scenario Studio session
  - name: createScenarioJob
    description: Create Scenario job
  - name: getScenarioJob
    description: Get Scenario job status
  - name: cancelScenarioJob
    description: Cancel Scenario job
  - name: listKnowledgeHubFolders
    description: List folders
  - name: getKnowledgeHubFolder
    description: Get folder
  - name: listKnowledgeHubPages
    description: List pages
  - name: createKnowledgeHubPage
    description: Create page
  - name: getKnowledgeHubPage
    description: Get page
  - name: updateKnowledgeHubPage
    description: Update page
  - name: replaceKnowledgeHubPage
    description: Update page (PUT)
  - name: archiveKnowledgeHubPage
    description: Archive page
  - name: listKnowledgeHubPageVersions
    description: List page versions
  - name: listKnowledgeHubSources
    description: List sources
  - name: createKnowledgeHubSource
    description: Create source
  - name: getKnowledgeHubSource
    description: Get source
  - name: deleteKnowledgeHubSource
    description: Delete source
rejected_hosts:
- host: https://mcp.exec.com
  status: 200
  reason: Wildcard DNS catch-all serving the exec.com marketing homepage (<title> "Exec – AI Training
    Platform..."). Not an MCP endpoint. Same false positive as trust.exec.com.
- host: https://api.exec.com/mcp
  status: 404
- host: https://www.exec.com/mcp
  status: 404
x-evidence:
  fetched: '2026-08-14'
  url: https://docs.exec.com/mcp
  http_status: 200
  content_type: text/event-stream
  methods_probed:
  - initialize
  - tools/list
  - resources/list
  - resources/read
  tool_count: 3
  resource_count: 1