Bastion

A local-first control plane for your AI agent's tools. Bastion is a gateway that sits between an AI agent (Claude Code, Cursor, Claude Desktop, …) and the MCP servers it uses — capping spend, rate-limiting, enforcing per-tool permissions, redacting or blocking dangerous arguments, and auditing every call.

One config file, one command, no cloud, no database.

Why

Model Context Protocol won — every major AI vendor ships it and there are 17,000+ MCP servers. But agents call those servers with no spend caps, no rate limits, no permissions, and no audit trail. A looping agent can burn real money in minutes, and you have no record of what it did.

Bastion is the missing control layer. One config file, one command, no cloud, no database.

How it works

Your agent points at Bastion instead of at its MCP servers directly. Bastion is both an MCP server (to the agent) and an MCP client (to the real "upstream" servers). It aggregates your upstreams behind one endpoint and enforces rules on every tools/call:

  AI agent  ──MCP──▶  Bastion  ──MCP──▶  upstream server A
                      (policy +          upstream server B
                       audit)            upstream server C

Install

pip install bastion-mcp

This installs the bastion command. To run it without installing globally, use uv:

uvx --from bastion-mcp bastion --help

To hack on Bastion itself, see CONTRIBUTING.md.

Quickstart

Create bastion.yaml:

upstreams:
  files:
    command: npx
    args: ["-y", "@modelcontextprotocol/server-filesystem", "/path/to/project"]

Validate it:

bastion validate

Point your agent at the gateway — e.g. for Claude Code:

claude mcp add --transport stdio bastion -- bastion run --config ./bastion.yaml

Your agent now reaches the filesystem server through Bastion. With a single upstream, tools keep their original names; configure several and Bastion namespaces each tool by its upstream key (files_read_file, search_query, …) so they never collide.

Configuration

bastion.yaml — upstreams is required; everything else is optional. A small example with permissions and audit:

upstreams:
  files:
    command: npx
    args: ["-y", "@modelcontextprotocol/server-filesystem", "/data"]

audit:
  enabled: true
  path: ./bastion-audit.jsonl

policy:
  default: allow
  permissions:
    - { tool: "files_read_*",   action: allow }
    - { tool: "files_delete_*", action: deny  }

A denied call is blocked before reaching the upstream and recorded in the audit log. Rate limits, budgets (call-count and cost), and argument guards (block or redact) layer onto the same policy section.

See [`docs/configuration.md`](docs/configuration.md) for the full reference of every section, [`docs/security.md`](docs/security.md) for the security model (secrets in the audit log, unauthenticated surfaces), and [`examples/`](examples/) for runnable starter configurations.

Dashboard

bastion dashboard serves a local web view of the audit log — every tool call, live, with arguments, outcomes, and timings:

bastion dashboard --config bastion.yaml   # then open http://127.0.0.1:8787

Milestones

Bastion shipped one feature per release, milestone by milestone:

• [x] M0 — multi-upstream proxy (stdio + HTTP), config schema, CLI
• [x] M1 — audit log: every tool call recorded to JSONL
• [x] M2 — permissions: per-tool allow/deny rules
• [x] M3 — first PyPI release (pip install bastion-mcp)
• [x] M4 — rate limiting: per-tool and global token buckets
• [x] M5 — budgets: call-count and cost caps with on-disk checkpoint
• [x] M6 — argument guards: regex-on-JSONPath, block or redact
• [x] M7 — live log viewer (bastion tail/logs/stats/init) + Windows CI
• [x] M8 — docs, examples, v1.0.0 🎉

Contributing

Issues and PRs welcome — see CONTRIBUTING.md.

License

Apache 2.0