# midplane

Safe-by-default SQL guardrails for AI agents. MCP server for Postgres over stdio + Streamable HTTP: parses every statement with a real SQL AST, enforces a per-table access policy, and audits before the query runs.

- **Type:** MCP server
- **Trust:** 85/100 (A), scored on the package rubric
- **Verification:** verified (build provenance)
- **Version:** 0.20.0
- **Author:** Unknown
- **License:** MIT
- **npm:** midplane
- **Source:** https://github.com/midplaneai/midplane
- **Compatible clients:** claude-code, cursor, copilot, gemini (basis: transport)

## Trust

85/100 (A), scored on the package rubric
- Publisher verified: no
- Build provenance: verified attestation
- npm trusted publishing (OIDC): yes
- Install scripts: nothing suspicious found
- Prompt-injection scan: not run
- Obfuscation scan: not run
- Evidence age: 36 days

## Security scan

- **Status:** clean
- **Scanned:** 2026-08-30T00:17:18.988Z
- **Version scanned:** 0.20.0
- **CVEs:** none found by OSV at scan time

## Tools

5 declared. Statically extracted from the shipped source — a floor on the surface, not a census.
- `query` — Parses the SQL with libpg_query, applies Midplane policy (table_access, multi_statement, tenant_scope, parse_error), audits the call, and returns rows on ALLOW.
- `list_tables` — Routed through the same policy + audit pipeline. Defaults to the 'public' schema.
- `describe_table` — Returns column name, data type, nullability, and default. Routed through the policy + audit pipeline.
- `list_databases` — Returns each database's name, whether tenant_scope is enforced, and its table_access default. Use this to discover which `database` values to pass to `query`, `
- `check_approval` — Poll a held write's approval status WITHOUT running anything. Use this to wait — it is cheap, safe to call repeatedly, and never consumes the approval. Returns 

## Install

**Verdict: install** — No blocking findings and no open coverage gaps — safe to install as configured.
**Config** (claude-code):
```json
"{\n  \"mcpServers\": {\n    \"midplane\": {\n      \"command\": \"npx\",\n      \"args\": [\n        \"-y\",\n        \"midplane\"\n      ],\n      \"env\": {\n        \"DATABASE_URL\": \"<YOUR_DATABASE_URL>\"\n      }\n    }\n  }\n}"
```
**Credentials it will ask for** (names only — Forge never holds a value):
- `DATABASE_URL` — Database URL (required)
Placeholders only. Forge never holds, brokers, or transmits a credential value — replace each <YOUR_NAME> in your own config file. Do not send a value back to Forge; no Forge endpoint accepts one.
- This entry needs 1 credential (1 required). The generated config carries placeholders, so it will fail in the editor rather than at runtime if they are left unset.

## Blast radius

Moderate blast radius — holds an api key; runs on your machine.
- Floor 23, ceiling 23 (tier: moderate)
- This is impact, not likelihood. A high radius is not a defect: a filesystem server is supposed to write files. It is never part of the trust score.

## Machine-readable views of this entry

- Signed JSON: https://forgeregistry.com/api/v1/packages/midplane
- Install plan: https://forgeregistry.com/api/v1/packages/midplane/install-plan
- Alternatives: https://forgeregistry.com/api/v1/alternatives/midplane
- HTML page: https://forgeregistry.com/registry/midplane
- MCP: POST https://forgeregistry.com/api/mcp → `forge_get_package` / `forge_install_plan`

## About this document

Generated by Forge (https://forgeregistry.com) — a compact rendering of the same record served, signed, at the JSON URL above. Trust and scan facts are the registry's own measurements; anything Forge did not measure is named as unmeasured rather than omitted.