{
  "markdown": "# Vector Decisions MCP Server\n\nMCP-native governance and action-assurance primitives for autonomous AI.\n\n## Tools\n\n- `gatri_trust_score` — legacy composite trust scoring.\n- `assure_action` — evidence-backed, context-aware assurance for one agent action.\n- `decide_action` — converts GATRI assurance + policy + permissions + context into `ALLOW`, `RESTRICT`, `REQUIRE_HUMAN`, or `DENY`.\n- `risk_assessment` — deployment risk assessment.\n- `compliance_check` — EU AI Act-oriented policy check.\n- `kill_switch` — emergency stop directive.\n\n## Core model\n\n```text\nagent × action × context × evidence\n                 ↓\n              GATRI\n                 ↓\n        Vector Decisions\n                 ↓\n      ALLOW / RESTRICT /\n      REQUIRE_HUMAN / DENY\n```\n\nThe important primitive is **action assurance**, not a generic claim that an agent is safe. The same agent can be acceptable for one action and unacceptable for another.\n\n## Local development\n\n```bash\nnpm install\nnpm run build\nnpm start\n```\n\nThe server uses stdio transport and is intended to be consumed by MCP-compatible clients.\n\n## Status\n\nVersion 1.1.0. This is a protocol-oriented V0: deterministic local evaluation first; evidence ingestion, persistent assurance records, policy registries, signed attestations and remote APIs are subsequent layers.\n",
  "bytes": 1312,
  "sha": "d66b1434e23a9a86169c6315cb264374e5a50bc61320c5990de6477597936054",
  "repo_slug": "nikitadatar/vectordecisions-mcp-server",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_nikitadatar_vectordecisions_mc_50f819de/readme"
}