{
  "markdown": "# VeriSwarm SDK & Integrations\n\n[![pip veriswarm](https://img.shields.io/pypi/v/veriswarm?label=pip%20veriswarm)](https://pypi.org/project/veriswarm/)\n[![pip veriswarm-mcp](https://img.shields.io/pypi/v/veriswarm-mcp?label=pip%20veriswarm-mcp)](https://pypi.org/project/veriswarm-mcp/)\n[![npm @veriswarm/sdk](https://img.shields.io/npm/v/@veriswarm/sdk?label=npm%20%40veriswarm%2Fsdk)](https://www.npmjs.com/package/@veriswarm/sdk)\n[![MCP Registry](https://img.shields.io/badge/MCP%20Registry-io.github.veriswarm%2Fveriswarm--mcp-1f6feb)](https://github.com/mcp/veriswarm/veriswarm-mcp)\n\nOfficial client libraries, MCP server, and plugins for [VeriSwarm](https://veriswarm.ai) — trust infrastructure for AI agents.\n\n## Packages\n\n| Package | Language | Install | Description |\n|---------|----------|---------|-------------|\n| [**Python SDK**](./python/) | Python | `pip install veriswarm` | REST client with Workflows, credential issuance, scoring profiles, LangChain adapter |\n| [**Node.js SDK**](./node/) | JavaScript | `npm install @veriswarm/sdk` | ESM client for decisions, events, Workflows, agent management |\n| [**MCP Server**](./mcp-server/) | Python | `pip install veriswarm-mcp` | 89 MCP tools for trust scoring, Guard, Passport, Vault, Workflows, Compliance (OWASP/EU AI Act/NIST/ISO 42001), Cedar policies + ABAC attributes, SRE, context governance, and content provenance (EU AI Act Art. 50). Works with Claude Desktop, Cursor, and any MCP client. (`approve_jit_grant` and `issue_jit_token` are deliberately not exposed to the LLM — those actions live behind an authenticated session in the web UI.) |\n| [**OpenClaw Plugin**](./openclaw-plugin/) | TypeScript | `openclaw plugins install veriswarm` | 11 tools + 3 hooks for OpenClaw agents. PII tokenization, policy enforcement, audit. Per-feature enable/disable. |\n| [**GitHub Action**](./github-action/) | Python | GitHub Marketplace | CI/CD trust gate — check agent trust scores in your pipeline |\n| [**JSON Schemas**](./schemas/) | JSON | `npm install @veriswarm/schemas` | Schema definitions for events, profiles, scores, and workflows |\n\n## Quick Start\n\n### MCP Server (Recommended)\n\nAdd to your MCP client config (Claude Desktop, Cursor, etc.):\n\n```json\n{\n  \"mcpServers\": {\n    \"veriswarm\": {\n      \"command\": \"python3\",\n      \"args\": [\"-m\", \"veriswarm_mcp\"],\n      \"env\": {\n        \"VERISWARM_API_URL\": \"https://api.veriswarm.ai\",\n        \"VERISWARM_API_KEY\": \"YOUR_API_KEY\"\n      }\n    }\n  }\n}\n```\n\nOr use the setup CLI:\n\n```bash\npip install veriswarm-mcp\nveriswarm-setup --platform claude --api-key YOUR_API_KEY\n```\n\nThis installs the MCP server, Guard hooks (PII protection + activity logging), and Guard Proxy config.\n\n### Python SDK\n\n```python\nfrom veriswarm import VeriSwarmClient\n\nclient = VeriSwarmClient(\"https://api.veriswarm.ai\", \"YOUR_API_KEY\")\n\n# Register an agent\nagent = client.register_agent({\"slug\": \"my-agent\", \"display_name\": \"My Agent\"})\n\n# Check if an action should be allowed\nresult = client.check_decision(agent_id=\"agt_123\", action_type=\"send_email\")\nprint(result[\"decision\"])  # \"allow\", \"review\", or \"deny\"\n\n# Ingest events\nclient.ingest_event(\n    event_id=\"evt_001\", agent_id=\"agt_123\", source_type=\"platform\",\n    event_type=\"task.completed\", occurred_at=\"2026-04-01T00:00:00Z\",\n    payload={\"task\": \"onboarding\", \"success\": True}\n)\n```\n\n### Node.js SDK\n\n```javascript\nimport { VeriSwarmClient } from '@veriswarm/sdk'\n\nconst client = new VeriSwarmClient({\n  baseUrl: 'https://api.veriswarm.ai',\n  apiKey: 'YOUR_API_KEY'\n})\n\nconst result = await client.checkDecision({\n  agentId: 'agt_123',\n  actionType: 'send_email'\n})\n```\n\n### OpenClaw Plugin\n\n```json5\n{\n  plugins: {\n    entries: {\n      veriswarm: {\n        enabled: true,\n        config: {\n          apiKey: \"YOUR_API_KEY\",\n          piiEnabled: true,\n          policyEnabled: true,\n          injectionScan: true,\n          auditEnabled: true\n        }\n      }\n    }\n  }\n}\n```\n\n## Free Plan\n\nGet started with no credit card:\n- 5,000 trust score queries/day\n- Up to 10 agents\n- Unlimited event ingestion\n- Portable credentials issued on demand (rate-limited per plan; see [pricing](https://veriswarm.ai/pricing) for current limits)\n\nSign up at [veriswarm.ai](https://veriswarm.ai/auth?mode=register).\n\n## Links\n\n- [VeriSwarm Platform](https://veriswarm.ai)\n- [Documentation](https://veriswarm.ai/docs)\n- [API Reference](https://veriswarm.ai/docs/api)\n- [Quickstart](https://veriswarm.ai/docs/quickstart)\n- [Pricing](https://veriswarm.ai/pricing)\n\n## License\n\nMIT\n",
  "bytes": 4516,
  "sha": "16a17ca7cbb7709149fa552583c77177d47c23dd46aa8c4f1d7084c72ac52b9d",
  "repo_slug": "veriswarm/veriswarm-sdk",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_veriswarm_veriswarm_mcp_d07e3458/readme"
}