# Agent Reliability Skills

> Agent Reliability Skills is a freemium MCP server for developer tools. Generate a pre-coding reliability skill for your agent repo from Trustabl's rules. It runs locally on your machine and is started by your AI client.

- **Kind:** MCP server
- **Pricing:** Freemium
- **Starting price:** $19.98 a month
- **Platforms:** CLI
- **Website:** https://trustabl.ai
- **Repository:** https://github.com/trustabl/agent-reliability-skills
- **MCP Registry name:** io.github.trustabl/agent-reliability-skills
- **Categories:** developer tools, security
- **Listing updated:** 2026-11-14

## About

Rather than scanning code after the fact, this server writes guidance for the agent that is about to write it. Its single tool, trustabl_forge, reads a repository's dependency manifests (package.json, pyproject.toml, requirements.txt, go.mod and others), finds which agent SDKs are used, and builds a SKILL.md from Trustabl's detection rules for exactly those SDKs.

The README lists the Claude Agent SDK, OpenAI Agents SDK, Google ADK, MCP, LangChain and LangGraph, CrewAI, Pydantic AI, the Vercel AI SDK and AutoGen. The skill is generated per repository and changes as dependencies and rules change. A file is written only when output_path is given, must stay inside the repo, and an existing file is replaced only if Forge generated it.

On first use it downloads the Trustabl analyzer release for your OS, verifies it against the release checksums and caches it locally. It needs Node.js 22 or later and is distributed as a GitHub release file, with an npm package and an mcpb bundle listed in the registry. The wider Trustabl service also sells scanning plans.

## Features

- **Per-repo skill generation:** Builds a SKILL.md tailored to the SDKs found in your dependencies.
- **SDK detection:** Reads package.json, pyproject.toml, requirements.txt, go.mod and similar manifests.
- **Safe file writes:** Writes only when output_path is set and never overwrites hand-written files.
- **Verified analyzer download:** Checks the analyzer against release checksums and caches it.
- **Policy and language overrides:** Add SDK categories or languages not yet in a manifest.
- **Workspace-aware:** Uses MCP roots when the client reports them, else a path.

## Good for

- Give a coding agent reliability rules before it writes agent code
- Regenerate the skill after upgrading agent SDKs
- Add guardrail and tool-safety rules to a Claude Code project
- Standardise agent safety guidance across repos
- Target rules for a new SDK before it appears in a manifest

## Install

### Claude Code (Terminal)

```sh
claude mcp add agent-reliability-skills -- npx -y @trustabl/agent-reliability-skills
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "agent-reliability-skills": {
      "command": "npx",
      "args": [
        "-y",
        "@trustabl/agent-reliability-skills"
      ]
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "agent-reliability-skills": {
      "command": "npx",
      "args": [
        "-y",
        "@trustabl/agent-reliability-skills"
      ]
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "agent-reliability-skills": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@trustabl/agent-reliability-skills"
      ]
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "agent-reliability-skills": {
      "command": "npx",
      "args": [
        "-y",
        "@trustabl/agent-reliability-skills"
      ]
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "agent-reliability-skills": {
      "command": "npx",
      "args": [
        "-y",
        "@trustabl/agent-reliability-skills"
      ]
    }
  }
}
```

## Tools

- `trustabl_forge`: Generates a pre-coding reliability SKILL.md for a repository.

## FAQ

### How many tools does it expose?

One, trustabl_forge.

### Which SDKs are supported?

Claude Agent SDK, OpenAI Agents SDK, Google ADK, MCP, LangChain, LangGraph, CrewAI, Pydantic AI, Vercel AI SDK and AutoGen.

### What does it need?

Node.js 22 or later; it downloads the analyzer on first use.

### Does it overwrite my files?

Only a skill file Forge generated, and only when you pass output_path.

### Is Agent Reliability Skills free?

Agent Reliability Skills has a free plan, with paid plans for more usage or features. Paid plans start at $19.98 a month.

### How do I add Agent Reliability Skills to Claude Code?

Run this command in your terminal: claude mcp add agent-reliability-skills -- npx -y @trustabl/agent-reliability-skills

### What are the best alternatives to Agent Reliability Skills?

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

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [GitHub](https://ai-powered.com/mcp/github-2): GitHub is an open-source MCP server for developer tools.
- [Serena MCP: the IDE for your agent](https://ai-powered.com/mcp/serena-mcp-the-ide-for-your-agent): Serena MCP: the IDE for your agent is an open-source MCP server for developer tools.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Ouroboros](https://ai-powered.com/mcp/ouroboros): Ouroboros is an open-source MCP server for developer tools.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [Concord MCP](https://ai-powered.com/mcp/concord-mcp): Concord MCP is a freemium MCP server for developer tools.

---
Source: https://ai-powered.com/mcp/agent-reliability-skills
