Skip to content
MCP server

Slipmark

Record AI agent verdicts bound to file hashes and refuse stale ones before shipping

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, self-hosted
Runs
Locally
Version
0.6.2
Listing updated
3 days ago

About Slipmark

Slipmark is an open-source MCP server for security. It runs locally on your machine and is started by your AI client.

Each verdict is tied to a SHA-256 hash of the exact files the agent read. Before anything ships, the gate re-hashes them and refuses on a hash mismatch, a missing file, a failing verdict, or no attestation at all. It is deterministic, makes no model calls and needs no account, and it runs as a GitHub Action. Slipmark says it does not judge whether an agent is right, only that the verdict is about the code being shipped.

The name comes from the paint line engineers mark across a tightened bolt. The authors say their own nightly pipeline of 20 or more agents had a security audit gate checking the wrong branch for months, and that pointed at 23 open agent pull requests the new gate refused all 23. That is an internal figure.

A second feature, drift diagnosis, codifies behavioural locks for AI features and, when a change breaks one, suggests what likely broke it, states how sure it is and hands the coding agent a repair prompt. Setup is npx slipmark init, which adds a report-only workflow and agent instructions. It is Apache-2.0 and from Reply Intelligence Group, formerly called DriftGuard, with an optional hosted Slipmark Cloud.

Install

Run in a terminal

claude mcp add slipmark -- npx -y slipmark

Add to claude_desktop_config.json

{
  "mcpServers": {
    "slipmark": {
      "command": "npx",
      "args": [
        "-y",
        "slipmark"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "slipmark": {
      "command": "npx",
      "args": [
        "-y",
        "slipmark"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "slipmark": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "slipmark"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "slipmark": {
      "command": "npx",
      "args": [
        "-y",
        "slipmark"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "slipmark": {
      "command": "npx",
      "args": [
        "-y",
        "slipmark"
      ]
    }
  }
}

What it does

Hash-bound verdicts
Binds each agent sign-off to a SHA-256 hash of the files it read.
Re-hash gate
Re-hashes files before shipping and refuses on any change.
Deterministic checks
No model calls, scoring or judgement, so results repeat.
Drift diagnosis
Codifies behavioural locks and explains likely causes with stated confidence.
One-command setup
npx slipmark init adds a report-only GitHub workflow.
Claude Code plugin
Installs through the plugin marketplace.

Good for

  • Stop a stale security-audit sign-off from shipping
  • Require that every spec file has a current passing verdict
  • Detect when a prompt or code change breaks an AI feature
  • Keep audit evidence for agent-written pull requests

Pricing

Open source

Open source under Apache-2.0; optional hosted Slipmark Cloud.

Prices change. Check replyintelligencegroup.co.uk before you buy.

Slipmark FAQ

Does it run models?

No. The verification is deterministic with no model calls.

What licence is it under?

Apache-2.0.

What does it not prove?

That the agent was right; it only proves the verdict was given on these exact bytes.

How do I set it up?

Run npx slipmark init in the repository, or add the GitHub Action step directly.

Is Slipmark free?

Yes. Slipmark is open source, so you can use and self-host it for free.

How do I add Slipmark to Claude Code?

Run this command in your terminal: claude mcp add slipmark -- npx -y slipmark

What are the best alternatives to Slipmark?

The closest alternatives to Slipmark are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

Alternatives to Slipmark

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools