{"_id":"@agentlaw/gateway","name":"@agentlaw/gateway","dist-tags":{"latest":"0.1.0"},"versions":{"0.1.0":{"name":"@agentlaw/gateway","version":"0.1.0","description":"The agentlaw CLI and MCP policy gateway: enforce a signed policy manifest on every tool call an AI agent makes.","type":"module","license":"Apache-2.0","engines":{"node":">=22"},"main":"./dist/index.js","types":"./dist/index.d.ts","bin":{"agentlaw":"dist/cli.js"},"exports":{".":{"types":"./dist/index.d.ts","import":"./dist/index.js"}},"scripts":{"build":"tsc -b","typecheck":"tsc -b","pretest":"tsc -b","dev":"tsx src/cli.ts","test":"node --import tsx --test \"test/**/*.test.ts\""},"dependencies":{"@agentlaw/audit":"0.1.0","@agentlaw/core":"0.1.0","@agentlaw/sink-supabase":"0.1.0","@modelcontextprotocol/sdk":"^1.25.2","yaml":"^2.7.0"},"publishConfig":{"access":"public","registry":"https://registry.npmjs.org/"},"repository":{"type":"git","url":"git+https://github.com/Mounaji-Webs/agentlaw.git","directory":"packages/gateway"},"homepage":"https://agentlaw.dev","bugs":{"url":"https://github.com/Mounaji-Webs/agentlaw/issues"},"keywords":["ai-agents","security","policy","mcp","governance","agentlaw","cli","gateway","model-context-protocol"],"gitHead":"a0e1b6f4b67936c9600daa13f63c1e45f8b9406b","_id":"@agentlaw/gateway@0.1.0","_nodeVersion":"24.18.0","_npmVersion":"11.16.0","dist":{"integrity":"sha512-kVrBEwf7cPT1ccRNI0UzL/6pl0i0FpSdATK4sdY29qJwAlzHTGhyHd1AQyh1hNRo6d5D+HgfviLAO4gzMREH1A==","shasum":"a06ff0df5d42a3fb35cc733a7beaa0b21543f0fd","tarball":"https://registry.npmjs.org/@agentlaw/gateway/-/gateway-0.1.0.tgz","fileCount":37,"unpackedSize":110194,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEUCIQDbrkFmq/Mf9OECe4h8Kyp6uxlOtvY9WLlj7OgfTFj69AIgQXzHtHKkKAq/noV3TttObeKIrrIw54PSuBVLpLqc3vQ="}]},"_npmUser":{"name":"lucilo","email":"lucilo@gmail.com"},"directories":{},"maintainers":[{"name":"lucilo","email":"lucilo@gmail.com"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/gateway_0.1.0_1785646053036_0.700481199246693"},"_hasShrinkwrap":false}},"time":{"created":"2026-08-02T04:47:32.844Z","0.1.0":"2026-08-02T04:47:33.164Z","modified":"2026-08-02T04:47:33.418Z"},"maintainers":[{"name":"lucilo","email":"lucilo@gmail.com"}],"description":"The agentlaw CLI and MCP policy gateway: enforce a signed policy manifest on every tool call an AI agent makes.","homepage":"https://agentlaw.dev","keywords":["ai-agents","security","policy","mcp","governance","agentlaw","cli","gateway","model-context-protocol"],"repository":{"type":"git","url":"git+https://github.com/Mounaji-Webs/agentlaw.git","directory":"packages/gateway"},"bugs":{"url":"https://github.com/Mounaji-Webs/agentlaw/issues"},"license":"Apache-2.0","readme":"# @agentlaw/gateway\n\nThe `agentlaw` CLI and MCP policy gateway — the enforcement point of [AgentLaw](https://github.com/Mounaji-Webs/agentlaw). Wraps any MCP server behind a signed policy manifest: every tool call is checked by the deterministic engine in `@agentlaw/core` and allowed, paused for human approval, or denied — fail-closed, hash-chain audited.\n\n## Quickstart\n\n```sh\nnpx @agentlaw/gateway wizard .\n```\n\nScaffolds a signed, ready-to-run `policy.yaml` (template + thresholds + keypair) and prints the `.mcp.json` snippet to wrap your MCP servers:\n\n```jsonc\n{\n  \"mcpServers\": {\n    \"safe-fs\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@agentlaw/gateway\", \"--policy\", \"policy.yaml\", \"--trust\", \"policy.pub\",\n               \"--\", \"npx\", \"-y\", \"@modelcontextprotocol/server-filesystem\", \"/sandbox\"]\n    }\n  }\n}\n```\n\n## CLI\n\n| Command | What it does |\n| --- | --- |\n| `agentlaw wizard .` | Guided setup: template, thresholds, signed policy + keypair. |\n| `agentlaw sign` | Sign a manifest with your Ed25519 private key. |\n| `agentlaw verify` | Verify a manifest's signature against a trusted public key. |\n| `agentlaw explain` | Dry-run: the verdict a tool call would receive, naming the rule. |\n| `agentlaw audit verify` | Check the integrity of the hash-chained audit log. |\n\nHuman-in-the-loop `approve` rules prompt on the enforcement host's own terminal — no external service; no terminal means deny, never silent allow.\n\nSpec and threat model: [RFC-0001](https://github.com/Mounaji-Webs/agentlaw/blob/main/spec/rfc-0001-policy-manifest.md).\n\nApache-2.0.\n","readmeFilename":"README.md","_rev":"1-8d42ff2af448b1751f79c98c1381aab4"}