PatchLabs

Working Security A

dev.patchlabs/patchlabsv3.1.0 Domain verified

Security checks for AI-built apps (Lovable, Supabase, Bolt): RLS, headers, monitoring, fixes.

Security

Watch Source

Registry entry

Open in the registry

{
  "_meta": {
    "io.modelcontextprotocol.registry/official": {
      "status": "active",
      "isLatest": true,
      "updatedAt": "2026-11-11T10:55:25.484273Z",
      "publishedAt": "2026-11-11T10:55:25.484273Z",
      "statusChangedAt": "2026-11-11T10:55:25.484273Z"
    }
  },
  "server": {
    "name": "dev.patchlabs/patchlabs",
    "title": "PatchLabs",
    "$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
    "remotes": [
      {
        "url": "https://api.patchlabs.dev/mcp",
        "type": "streamable-http",
        "headers": [
          {
            "name": "X-API-Key",
            "isSecret": true,
            "description": "Optional PatchLabs API key (pl_live_...) for the app tools. The free rls_check and headers_check tools work without one."
          }
        ]
      }
    ],
    "version": "3.1.0",
    "repository": {
      "url": "https://github.com/manra399/patchlabs",
      "source": "github"
    },
    "description": "Security checks for AI-built apps (Lovable, Supabase, Bolt): RLS, headers, monitoring, fixes."
  }
}

Maintain this server? Claim it

Try it

Call this server's tools from your browser. Nothing to install.

https://api.patchlabs.dev/mcp

X-API-Key (optional)

Optional PatchLabs API key (pl_live_...) for the app tools. The free rls_check and headers_check tools work without one.

Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.

Tool list from our last check. Connect live to refresh it.

get_check

Status, summary, report and fix for a check id.

check_idstring, required

This tool isn't marked read-only, so it may change something on the server or in an account you connected.

I understand, run it anyway

Install

Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.

One click: Add to Cursor Add to VS Code Add to LM Studio Add to Goose

Run in a terminal

claude mcp add --transport http patchlabs https://api.patchlabs.dev/mcp

Headers

X-API-Key (optional, not in the config above): Optional PatchLabs API key (pl_live_...) for the app tools. The free rls_check and headers_check tools work without one.

Tools (6)

get_check

Status, summary, report and fix for a check id.

Argument	Type	Description
check_id*	string

Input schema (JSON)

{
  "type": "object",
  "required": [
    "check_id"
  ],
  "properties": {
    "check_id": {
      "type": "string"
    }
  },
  "additionalProperties": false
}

headers_check

Grade a public site's HTTPS, TLS certificate, security headers and cookie flags from one load of its home page (no sign-up). Returns findings PL-027..PL-036 with fixes.

Argument	Type	Description
url*	string	A public site, e.g. https://app.example.com

Input schema (JSON)

{
  "type": "object",
  "required": [
    "url"
  ],
  "properties": {
    "url": {
      "type": "string",
      "description": "A public site, e.g. https://app.example.com"
    }
  },
  "additionalProperties": false
}

list_apps

Apps monitored in the API key's workspace, with their last check. Needs X-API-Key.

No arguments.

Input schema (JSON)

{
  "type": "object",
  "required": [],
  "properties": {},
  "additionalProperties": false
}

rls_check

Scan a public repo's supabase/migrations for row-level-security mistakes (static analysis, no sign-up).

Argument	Type	Description
repo*	string	Public GitHub repo, e.g. owner/name or https://github.com/owner/name

Input schema (JSON)

{
  "type": "object",
  "required": [
    "repo"
  ],
  "properties": {
    "repo": {
      "type": "string",
      "description": "Public GitHub repo, e.g. owner/name or https://github.com/owner/name"
    }
  },
  "additionalProperties": false
}

run_check

Queue a check for a monitored app now. Needs X-API-Key.

Argument	Type	Description
app_id*	string

Input schema (JSON)

{
  "type": "object",
  "required": [
    "app_id"
  ],
  "properties": {
    "app_id": {
      "type": "string"
    }
  },
  "additionalProperties": false
}

start_check

Request the full lab check. Returns a check id and a link; the repo owner proves ownership with a one-line file.

Argument	Type	Description
repo*	string	Public GitHub repo, e.g. owner/name or https://github.com/owner/name
email*	string	Where the report goes

Input schema (JSON)

{
  "type": "object",
  "required": [
    "repo",
    "email"
  ],
  "properties": {
    "repo": {
      "type": "string",
      "description": "Public GitHub repo, e.g. owner/name or https://github.com/owner/name"
    },
    "email": {
      "type": "string",
      "description": "Where the report goes"
    }
  },
  "additionalProperties": false
}

Security checkA100/100

We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 11, 2026.

Hidden or reordered text
Overriding the model's instructions
Hiding actions from the user
Reading keys or credentials
Fake system tags
Steering other servers' tools
Pulling in the conversation
Smuggling parameters
Demanding to be used for everything
Steering away from web search
Download-and-run commands
Encoded blobs
Invisible characters
Credentials as tool arguments
Very long descriptions
Read and write hintsNote1 found
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
- server · annotations
  6 of 6 tools: rls_check, headers_check, start_check, get_check, list_apps, ...
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.

How the security check works