@antops/mcp
OfficialClick on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@@antops/mcpwhat's the current change risk status for my workspace?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
AntOps Developer Tools
Official CLI, Python SDK, TypeScript SDK, GitHub Action and MCP server for the AntOps production API. These tools are thin clients: AntOps remains the source of truth for all product logic, scoring, monitoring and policy decisions.
Quick start
python -m pip install ./packages/python
export ANTOPS_API_KEY='your workspace key'
antops auth status
antops company lookup GB 00000006 --jsonUse ANTOPS_BASE_URL only for a compatible non-production endpoint. It defaults to
https://api.antops.dev. Never commit a key. API documentation is available at
https://api.antops.dev/docs.
Related MCP server: FolderBridge MCP
Packages
packages/python:antopsPython client and CLI.packages/typescript:@antops/sdkTypeScript client.actions/change-risk: reusable GitHub Action for deterministic Change Risk gating.packages/mcp:@antops/mcp, a safe stdio MCP server.
Security
CLI keys are read from ANTOPS_API_KEY or an explicit antops auth login --stdin config file.
They are never CLI flags. All clients use timeouts, bound response sizes and omit API keys from
errors. The Action reads a bounded list of static text files and never executes repository content.
MCP tools expose no destructive operation.
The MCP server also exposes bounded status/history reads for Change Risk, private documents, workspace overview and integration configuration. Each MCP request includes only a fixed allow-listed tool identifier for aggregate platform telemetry; it never forwards prompts, API keys or tool arguments as telemetry.
Releases
Each publishable package starts at 0.1.0 and follows semantic versioning. CI validates release
artifacts. Publishing is deliberately disabled until PyPI/npm trusted publishing or maintainer-owned
credentials are configured.
This server cannot be deployed
Maintenance
Related MCP Connectors
Run verified read-only code tools: quant diagnostics + agent-ops preflight, no source exposure.
Read-only smart-contract security intelligence for autonomous agents.
Sandbox workspace tools: search, file read, DB queries, integrations. Returns synthetic data.
Read-only finance and operations controls for AI agents with evidence and safe next actions.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceProvides read-only access to OpenClaw and Codex operational data like status, tasks, sessions, logs, and Git changes. Ensures safety via redaction, allowlists, and bounded outputs.MIT
- AlicenseAqualityAmaintenanceEnables AI clients to inspect and carefully edit a bounded local workspace via stdio, with a read-only default, conflict-safe updates, and no network listener or telemetry.9Apache 2.0
- FlicenseAqualityAmaintenanceEnables coding agents to perform workspace-confined file operations, read-only Git inspection, and structured shell commands, while requiring out-of-band human approval for mutations and external executions and maintaining an audit trail.143-
- AlicenseAqualityBmaintenanceEnables an AI agent to run local tools on the user's own machine via stdio, including command execution, workspace file read/write, and system status checks.52MIT