mcpField
S
safetyreplay-mcp

safetyreplay-mcp

Remote
github.com/clauxelAI & Agents0Updated 23w ago
Repository

About safetyreplay-mcp

A hosted remote MCP for replaying AI agent safety evaluations and gating releases. It supports policy checks, evaluation receipts, control-fix suggestions, and release evidence exports.

Key features

  1. 01Replay safety evaluations
  2. 02Check policy gates
  3. 03Record evaluation receipts
  4. 04Suggest control fixes
  5. 05Export release evidence

Use cases

  1. 01Replay safety checks before an agent release
  2. 02Check a release against policy gates
  3. 03Export evidence for a release review

MCP support

Exposes safety replay, policy-gate, receipt, control-fix, and evidence-export tools to MCP clients via a remote endpoint. Production access requires a bearer token.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://safetyreplay.clauxel.com/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

All endpoints
Streamable HTTPhttps://safetyreplay.clauxel.com/mcpAuth required

Tools

Details

From the project README

Safety Replay Gate MCP

Replay safety evals before agent releases move forward.

Paid remote MCP for AI agent safety replay checks, policy gates, eval receipts, control-fix suggestions, and release evidence exports.

Public Endpoints

Access

This is a paid hosted remote MCP. Production calls require a bearer token issued from the product website.

Authorization: Bearer <token>

Unauthenticated browser visits to /mcp return a clear JSON error instead of internal details.

Tools

  • run_safety_replay
  • check_policy_gate
  • record_eval_receipt
  • suggest_control_fix
  • export_release_evidence

Quick Start

  1. Open the website and choose a plan.
  2. Create or request an API token.
  3. Add the endpoint to an MCP client that supports Streamable HTTP remote servers.
  4. Send JSON-RPC requests with the bearer token.

Useful Links

Status

This repository is a public documentation and directory-submission reference for the hosted service. It does not contain the private production source code.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.