{
  "markdown": "<div align=\"center\">\n  <img src=\"docs/assets/rule-drift-mark.svg\" width=\"88\" height=\"88\" alt=\"RuleDrift logo\">\n  <h1>RuleDrift</h1>\n  <p><strong>Catch the moment your agent drifts from its rules.</strong></p>\n</div>\n\nRuleDrift tests whether an AI agent still follows important instructions after a conversation becomes long or distracting.\n\nRun it like a test suite when you change a prompt, model, memory system, or agent workflow. It runs locally and does not sit in front of production requests.\n\n## Quick start\n\nRequires Node.js 22+ and [Ollama](https://ollama.com/).\n\n```bash\nollama pull qwen3:1.7b\nnpx rule-drift init\nnpx rule-drift test\n```\n\nNo paid model API, account, backend, or telemetry is required.\n\n## What you get\n\n- The first conversation point where a rule fails\n- Terminal, JSON, and JUnit reports\n- Local Ollama and bring-your-own-agent support\n- Optional MCP tools for AI coding clients\n- Stable exit codes for CI\n\n## MCP\n\n```bash\nnpx -y rule-drift mcp --root /path/to/your/project\n```\n\nSee the [MCP setup guide](docs/mcp.md) for client configuration and safety details.\n\n## Documentation\n\n- [Full guide](docs/guide.md)\n- [All documentation](docs/README.md)\n\nContributions are welcome—see [CONTRIBUTING.md](CONTRIBUTING.md). Licensed under [MIT](LICENSE).\n",
  "bytes": 1276,
  "sha": "305b5d8ff9c26cd1488bca15ead9697b5ff99e627cb8ae440d0904bbb106014c",
  "repo_slug": "ri7in/rule-drift",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_ri7in_rule_drift_b4642100/readme"
}