Synced from monorepo Changes: - Workspace server: surface preview-proxy metrics through the hub metric pump - Shell: reclaim a session’s retained state in one entry - Shell: reclaim a session’s resident state in one entry - Pager: withhold key event types from Alacritty builds that double keys - Tools: cancel a session’s subagents when it closes - Pager: keep the whole plan in scrollback and separate reasoning from output in minimal mode - Pager: probe terminal version over DA2 and include it with feedback - SuperGrok Plus: identity, CLI, and analytics tier surfaces - Shell: inherit the session process scope into subagents - Pager: build @-file-search matcher lazily on first use - Tools: fix description and output contradictions in tool definitions - Workspace: degrade @-file-search instead of aborting on thread exhaustion - Tools: reap a session’s LSP servers when it closes - Tools: fix contradictions and defects in tool descriptions, schemas, and harness pools - MCP: reap stdio MCP children on session close - Shell: reuse spawn-time skill discovery for session telemetry - Tools: stop leaking shell-wrapper positional params into sourced scripts (fixes activate_conda under persistent/static shell) - Shell: self-heal corrupt session-search SQLite cache - Workspace: cap workspace-server tokio workers on many-core hosts - Shell: reap a session’s child processes when it closes - Crash handler: capture SIGABRT so panic-aborts leave crash reports - CLI chat proxy: team-scoped Grok Code managed-config admin routes - MCP: add CLI enable/disable for MCP servers - Shell: cap tokio worker threads for startup thread demand - Workspace: harden git_commit and add git_sync_base operation - Circuit breaker: add feature-gated gRPC retry policy Source-Revision: 2a818575225183d8ca915f5632a09b8067b5156a
16 KiB
MCP Servers
MCP (Model Context Protocol) servers extend Grok with external tool integrations. They let Grok interact with any service that implements the MCP standard.
What Are MCP Servers?
An MCP server is a process that exposes tools to Grok over a standardized protocol. When you configure an MCP server, its tools become available to the model alongside Grok's built-in tools. The model can discover and call these tools during a session.
For example, a GitHub MCP server might expose tools like create_issue, list_pull_requests, and search_code. A database server might expose query, list_tables, and describe_schema.
See the MCP specification for protocol details.
Configuration
MCP servers are configured in ~/.grok/config.toml under [mcp_servers.<name>] sections.
To distribute MCP servers to a team, or to restrict which servers users may run, see Distribute across an organization in the Plugins guide.
stdio Transport (Local Process)
Grok spawns a local process and communicates over stdin/stdout:
[mcp_servers.my-server]
command = "/path/to/server" # Server executable
args = ["--flag", "value"] # Command arguments
env = { API_KEY = "sk-..." } # Environment variables
enabled = true # Enable or disable the server (default: true)
startup_timeout_sec = 30 # Server startup timeout, seconds (default: 30)
tool_timeout_sec = 6000 # Per-tool-call timeout fallback, seconds (default: 6000)
tool_timeouts = { slow_op = 120 } # Per-tool timeout overrides, seconds
Global startup-timeout override: instead of setting
startup_timeout_secper server, you can change the default for all servers via theMCP_TIMEOUTenvironment variable (milliseconds, compatible with Claude Code) orGROK_MCP_STARTUP_TIMEOUT_SECS(seconds). A per-serverstartup_timeout_secstill takes precedence over both. Cold-startnpx/uvxservers that download packages on first launch often need this; the default is 30s.MCP tool-result size cap: large MCP /
use_toolresults are truncated inline (full payload spilled under the sessionmcp/folder). Default is 20_000 bytes. Override via:
- env
GROK_MAX_MCP_OUTPUT_BYTESorMAX_MCP_OUTPUT_BYTES(bytes; Grok-native wins if both set; Claude-style name, but we bound by bytes not tokens)config.toml— user-level (~/.grok/config.toml) or repo-level (.grok/config.tomlanywhere on the cwd → git-root chain; the deepest file wins, and the repo value applies only once the folder is trusted):[mcp] max_output_bytes = 40000Precedence: requirements.toml > env > repo
.grok/config.toml> user/managed config > default. Repo edits apply to running sessions in that directory via config hot-reload.
HTTP/SSE Transport (Remote Server)
For remote MCP servers accessible over HTTP:
[mcp_servers.remote-api]
url = "https://mcp.example.com/api"
headers = { "Authorization" = "Bearer token" }
Streamable HTTP with Session ID
[mcp_servers.my-streamable-server]
url = "https://mcp.example.com/api/mcp"
headers = { "x-mcp-session-id" = "{{session_id}}" }
CLI Management
Manage MCP servers from the command line without editing config files:
# List configured MCP servers
grok mcp list
grok mcp list --json # Machine-readable output
# Add a stdio server. Everything after -- is the server command, so flags
# like -y reach the server instead of being parsed by grok.
grok mcp add filesystem -- npx -y @modelcontextprotocol/server-filesystem /path/to/dir
# Add a stdio server with environment variables (-e is repeatable)
grok mcp add postgres -e DATABASE_URL=postgres://localhost/mydb -- npx -y @modelcontextprotocol/server-postgres
# Add a remote HTTP server
grok mcp add --transport http sentry https://mcp.sentry.dev/mcp
# Add a remote server with an authentication header (--header is repeatable)
grok mcp add --transport http api https://mcp.example.com/mcp --header "Authorization: Bearer YOUR_TOKEN"
# Add a remote SSE server
grok mcp add --transport sse linear https://mcp.linear.app/sse
# Remove a server
grok mcp remove github
# Enable or disable a local/TOML (or compat-sourced) server
grok mcp enable github
grok mcp disable github
# Diagnose a server's configuration and connectivity
grok mcp doctor # Check every configured server
grok mcp doctor github # Check one server
grok mcp doctor --json # Machine-readable output
The transport defaults to stdio; pass --transport http or --transport sse for remote servers.
By default grok mcp add writes to ~/.grok/config.toml (--scope user). Use --scope project to write to .grok/config.toml in the current directory instead, which can be committed and shared with your team (see Project-Scoped MCP Servers). Header and environment variable values are stored verbatim, so reference secrets as ${VAR} instead of pasting them into a committed project config (see Example Configurations). grok mcp list shows servers from both scopes, marking project-scoped ones with (project) and disabled ones with (disabled).
grok mcp remove searches both scopes and exits 0 after removing the server. It exits 1 when the name is not found, or when the name is defined in both user and project scope — pass --scope to say which one to remove.
grok mcp enable / disable persist the personal on/off state to user ~/.grok/config.toml (disabled_mcp_servers, and [mcp_servers.<name>].enabled when that entry exists). Scope:
- Known names: user/project Grok TOML, names already on the disabled list, compat sources (
.mcp.json, Claude, Cursor), plugin MCP servers (same discovery as doctor//mcps), and legacy managedgrok_com_*(no local entry required). - Enable only: if the cwd-nearest project definition has sticky
enabled = false, that single key is cleared (comments preserved); disable never rewrites project configs. - Not full
/mcpsparity: gateway connectors (managed_gateway:…, stored underdisabled_mcp_tools.__managed_gateway_connectors) stay Space-only in the TUI. Idempotent; unknown names exit 1.
Breaking changes from earlier releases: --env now takes one KEY=value per flag (use -e A=1 -e B=2, not --env A=1 B=2), and server names may only contain letters, numbers, hyphens, and underscores.
Project-Scoped MCP Servers
MCP servers can be configured per-project by placing a .grok/config.toml in your repository:
my-project/
.grok/
config.toml
src/
...
# .grok/config.toml
[mcp_servers.linear]
url = "https://mcp.linear.app/mcp"
enabled = true
When a server exposes a native HTTP/SSE endpoint, prefer the url form over wrapping it in a stdio proxy such as npx mcp-remote <url>. Grok handles HTTP/SSE and OAuth directly, so the native form avoids an extra subprocess per session. It also registers Grok's own OAuth client with the provider.
Grok walks from the current directory up to the git repo root, loading .grok/config.toml at each level:
| Location | Scope | Priority |
|---|---|---|
~/.grok/config.toml |
All projects | Lowest |
<repo-root>/.grok/config.toml |
This repository | Medium |
<cwd>/.grok/config.toml |
Current directory | Highest |
If a project defines a server with the same name as a global one, the project version replaces it entirely (fields are not merged).
Project-scoped files contribute [mcp_servers], [plugins], and [permission] entries. Grok reads most other config sections only from ~/.grok/config.toml.
Tool Naming
MCP tools are namespaced with the server name to avoid collisions:
- Server
filesystemwith toolread_filebecomesfilesystem__read_file - Server
githubwith toolcreate_issuebecomesgithub__create_issue
Toggle Servers at Runtime
You can enable or disable MCP servers without restarting Grok (TUI /mcps or CLI — see CLI Management).
The /mcps Modal
Open the MCP servers modal in the TUI:
- Run
/mcpsas a slash command - Or press
Ctrl+L(non–VS Code family) and navigate to the MCP Servers tab; on VS Code family use/pluginsor/mcpand open the MCP Servers tab
From the modal you can:
- See each server's source, enabled state, and tool count
- Enable or disable a server with
Space - Expand a server to view the tools it provides
- Refresh the list with
rafter you editconfig.toml - Authenticate an OAuth server with
i - Add a server with
a, or remove a local server withx(the modal asks for confirmation; press lowercaseyto remove, or any other key to cancel)
Tool Discovery
The model has access to two built-in tools for working with MCP servers:
search_tool— Discover available integration tools across all enabled MCP servers. Use this to find tools by name or description.use_tool— Call an integration tool discovered viasearch_tool. Specify the fully-qualified tool name (e.g.,github__create_issue).
Compatibility
Grok loads MCP server configurations from multiple sources for compatibility:
| Source | Format | Location | Configurable |
|---|---|---|---|
config.toml |
Native Grok config | ~/.grok/config.toml, .grok/config.toml |
Always on |
.claude.json |
Claude Code format | ~/.claude.json |
[compat.claude] mcps |
.cursor/mcp.json |
Cursor format | ~/.cursor/mcp.json, <project>/.cursor/mcp.json |
[compat.cursor] mcps |
.mcp.json |
MCP standard format | Project root (cwd to git root) | Loaded unless you have imported or dismissed the Claude import prompt (the import marker is set) |
All sources are merged in priority order: config.toml > Claude > Cursor > .mcp.json. Servers from higher-priority sources take precedence when names conflict.
The Claude and Cursor MCP sources are scanned by default. To disable scanning for a specific vendor, set [compat.<vendor>] mcps = false in ~/.grok/config.toml or the corresponding environment variable (GROK_CURSOR_MCPS_ENABLED, GROK_CLAUDE_MCPS_ENABLED). See Configuration for details. Use grok inspect to see which MCP servers were loaded and their vendor origin ([cursor], [claude]).
MCP OAuth
For MCP servers that require OAuth authentication, Grok handles the credential flow automatically. When an MCP server requests OAuth credentials, Grok opens a browser-based authorization flow and stores the resulting tokens for future use.
Example Configurations
Use the url form for hosted MCP servers and the command / args form for local stdio tools.
Native HTTP (hosted services)
You must authenticate OAuth-based MCP servers before you can use them. Grok stores the resulting tokens under ~/.grok/mcp_credentials.json as local plaintext with owner-only file permissions (0600 on Unix). Prefer full-disk encryption on the host. After you edit config.toml, press r in the /mcps modal to refresh the server list.
[mcp_servers.linear]
url = "https://mcp.linear.app/mcp"
enabled = true
[mcp_servers.sentry]
url = "https://mcp.sentry.dev/mcp"
enabled = true
[mcp_servers.mixpanel]
url = "https://mcp.mixpanel.com/mcp"
enabled = true
For internal or self-hosted servers that authenticate with a static bearer token rather than OAuth, set the Authorization header explicitly:
[mcp_servers.internal-tools]
url = "https://mcp.internal.example.com/mcp"
enabled = true
[mcp_servers.internal-tools.headers]
Authorization = "Bearer <token>"
To avoid putting secrets in the config file, reference an environment variable with ${VAR} (or ${VAR:-default}). Grok expands string fields in [mcp_servers.*] — url, command, args, and the values in env and headers — at load time:
[mcp_servers.internal-tools]
url = "https://mcp.internal.example.com/mcp"
enabled = true
headers = { "Authorization" = "Bearer ${INTERNAL_MCP_TOKEN}" }
Local stdio
Use stdio for tools that must run locally (filesystem access, local databases, in-house servers).
# Filesystem access scoped to a directory
[mcp_servers.filesystem]
command = "npx"
args = ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/allowed/directory"]
# Local Postgres
[mcp_servers.postgres]
command = "npx"
args = ["-y", "@modelcontextprotocol/server-postgres", "postgresql://user:pass@localhost/db"]
# Custom server with a longer startup timeout and tuned per-tool timeouts
[mcp_servers.my-tools]
command = "/usr/local/bin/my-mcp-server"
args = ["--config", "/etc/my-mcp.json"]
startup_timeout_sec = 30
tool_timeout_sec = 120
tool_timeouts = { slow_analysis = 300, quick_lookup = 10 }
On Windows, npm installs launchers like npx, npm, pnpm, and yarn as .cmd batch shims (there is no npx.exe). Grok resolves a bare command such as npx to its real launcher path on PATH (honoring PATHEXT) before spawning, so these work without manually wrapping them in cmd /c. A command given as an absolute path or one containing a path separator is used as-is.
Available MCP Servers
A partial list of MCP servers you can configure with the url or command forms shown above. Confirm the current endpoint or package name with each provider before use:
| Server | Transport | Endpoint / Package |
|---|---|---|
| Linear | HTTP (OAuth) | https://mcp.linear.app/mcp |
| Sentry | HTTP (OAuth) | https://mcp.sentry.dev/mcp |
| Mixpanel | HTTP (OAuth) | https://mcp.mixpanel.com/mcp |
| Filesystem | stdio | @modelcontextprotocol/server-filesystem |
| Git | stdio | @modelcontextprotocol/server-git |
| GitHub | stdio | @modelcontextprotocol/server-github |
| GitLab | stdio | @modelcontextprotocol/server-gitlab |
| PostgreSQL | stdio | @modelcontextprotocol/server-postgres |
| SQLite | stdio | @modelcontextprotocol/server-sqlite |
| Puppeteer | stdio | @modelcontextprotocol/server-puppeteer |
See the MCP Server Registry for the full list of community servers and the MCP specification for protocol details.
Subagents and MCP
Subagents inherit the parent session’s connected MCP servers by default, including plugin-sourced agents. Use agent frontmatter mcpInheritance to restrict that set (all, none, named, or except). Details are in Subagents — MCP inheritance.
If a child lists search_tool / use_tool but returns an empty catalog, check that:
- The parent session actually connected the server (see Extensions /
grok inspect) - The agent’s
mcpInheritanceis notnoneor a filter that excludes the server - Plugin agents cannot declare their own
mcpServersin frontmatter — they only see parent-connected servers
Troubleshooting
Server Not Starting
# Test the server command manually
npx -y @modelcontextprotocol/server-filesystem /path
# Increase startup timeout
# In config.toml:
[mcp_servers.filesystem]
startup_timeout_sec = 30
For stdio servers, Grok captures the process's standard error to ~/.grok/logs/mcp/<server>.stderr.log, truncated on each launch. Check this file when a server starts but fails to handshake:
tail -f ~/.grok/logs/mcp/filesystem.stderr.log
Viewing Server Status
Use grok inspect to see all loaded MCP servers and their sources:
grok inspect # Human-readable
grok inspect --json # Machine-readable
Debug Logging
RUST_LOG=debug GROK_LOG_FILE=/tmp/grok.log grok
tail -f /tmp/grok.log
Look for log entries containing mcp to trace server startup, tool discovery, and tool call execution.