{ "markdown": "

\n

@palveron/mcp-server

\n

Official Palveron MCP server — advisory AI-governance checks for MCP hosts and coding agents

\n

\n\n

\n \"npm\n \"npm\n \"License:\n \"Documentation\"\n \"Early\n

\n\n---\n\n> **⚡ Early access.** Palveron is launching soon and access is currently invite-gated. This server is fully functional — you just need a `pv_live_` key, which comes with an Early Access invite. **[Join the waitlist → palveron.com/early-access](https://palveron.com/early-access)**\n\nGive your MCP-capable client — Claude Code, Cursor, any other MCP host — governance tools it can call **before** executing an action: policy verdicts, PII-masking results, and an audit trace for every check, all from the Palveron AI Governance Gateway.\n\n- **Drop-in MCP server** — one binary, configure your MCP host, you're done\n- **Advisory governance verdicts** — the gateway decides allow / mask / deny / hold; the agent (or your wrapping code) acts on the verdict\n- **Audit trail per check** — trace IDs flow back to the Palveron dashboard\n- **Built on `@palveron/sdk`** — retry, circuit breaker, typed errors out of the box\n\n## Advisory check vs. enforced governance — which one do you need?\n\nThis package is the **advisory** path. It cannot physically stop a tool call: it answers the question \"is this allowed?\", and the calling agent (or your code via `check()`/`wrap()`) is expected to honor the answer. That is by design — it works with any MCP host, needs no network topology change, and gives you verdicts plus audit traces immediately.\n\n**Enforced, non-bypassable governance** for MCP tool calls is a different product surface: the **Palveron remote MCP proxy**. There you register the *target* MCP server in the Palveron dashboard and point your IDE at `POST {gateway}/api/v1/mcp/proxy/{server_id}` instead of the target — every `tools/call` then structurally passes through policy checks, approval holds, and drift detection before it can reach the target server. No cooperation from the agent required.\n\n| | This package (`@palveron/mcp-server`) | Remote MCP proxy |\n|---|---|---|\n| Model | Voluntary pre-flight check | In-path enforcement |\n| Can a call bypass it? | Yes — enforcement depends on the caller honoring the verdict | No — the proxy sits between client and target server |\n| Setup | `npx`, one env var | Register the target server in the dashboard, point the IDE at the proxy URL |\n| Use it when | You want verdicts + audit traces inside any MCP host, or wrap tool calls programmatically | You need governance that agents cannot skip |\n\nSee the [MCP setup guide](https://docs.palveron.com/mcp/setup) for the enforced path.\n\n## Installation\n\n```bash\nnpm install -g @palveron/mcp-server\n# or use directly via npx\nnpx @palveron/mcp-server\n```\n\n## Quick Start\n\n> The `pv_live_…` key in the examples below requires an **Early Access** invite — [request one here](https://palveron.com/early-access). Once you have a key, everything works out of the box.\n\n### Claude Code\n\nAdd to your `claude_desktop_config.json`:\n\n```json\n{\n \"mcpServers\": {\n \"palveron\": {\n \"command\": \"npx\",\n \"args\": [\"@palveron/mcp-server\"],\n \"env\": {\n \"PALVERON_API_KEY\": \"pv_live_xxx\",\n \"PALVERON_BASE_URL\": \"https://gateway.palveron.com\"\n }\n }\n }\n}\n```\n\n### Other MCP hosts\n\nAny MCP-capable host can launch the server via the `palveron-mcp` binary:\n\n```bash\nPALVERON_API_KEY=pv_live_xxx palveron-mcp\n```\n\n## Tools exposed via MCP\n\n| Tool | Purpose |\n|------|---------|\n| `palveron_check` | Check a tool call against governance policies before execution. Returns the decision (`PASSED`, `BLOCKED`, `MODIFIED`/`REDACTED`/`ANONYMIZED` with the sanitized input, `PENDING_APPROVAL`, …), whether the call may proceed, findings, and the audit trace ID. |\n| `palveron_status` | Governance status and diagnostics — gateway connectivity, circuit-breaker state, configured local lists. |\n| `palveron_policy_list` | List the project's active policies so the host can explain its decisions. |\n\n## Configuration\n\nThe server reads its configuration from environment variables. The MCP host sets them; you never put secrets in code.\n\n| Variable | Required | Description |\n|----------|:--------:|-------------|\n| `PALVERON_API_KEY` | yes | API key (`pv_live_…`) — comes with an [Early Access invite](https://palveron.com/early-access) |\n| `PALVERON_BASE_URL` | no | Override the gateway endpoint (default: `https://gateway.palveron.com`) |\n| `PALVERON_FAIL_OPEN` | no | `true` = allow tool calls when the gateway is unreachable (default: `false`, fail-closed) |\n| `PALVERON_LOG_LEVEL` | no | `debug` / `info` / `warn` / `error` / `silent` (default: `info`) |\n| `PALVERON_ALWAYS_BLOCK` | no | Comma-separated tool names that are always denied locally, without a gateway call |\n| `PALVERON_ALWAYS_ALLOW` | no | Comma-separated tool names that always pass locally, without a gateway call |\n\n## Requirements\n\n- Node.js **18 or newer**\n- An MCP-capable client (Claude Code 0.4+, Cursor, etc.)\n\n## Links\n\n- **Early Access / Waitlist** — [palveron.com/early-access](https://palveron.com/early-access)\n- **Documentation** — [docs.palveron.com/integrations/mcp](https://docs.palveron.com/integrations/mcp)\n- **Dashboard** — [palveron.com](https://palveron.com)\n- **Support** — [hello@palveron.com](mailto:hello@palveron.com)\n- **GitHub** — [palveron/mcp-server](https://github.com/palveron/mcp-server)\n- **Changelog** — [CHANGELOG.md](https://github.com/palveron/mcp-server/blob/main/CHANGELOG.md)\n\n## License\n\n[MIT](./LICENSE) — Copyright © 2026 Palveron.\n", "bytes": 6396, "sha": "f7d8ffc15a6d9da5cb78680b60dcf1b1c04af2266549f4585220b4491a4583a2", "repo_slug": "palveron/mcp-server", "fonte": "repo", "truncated": false, "api": "https://agentalog.com/api/listings/mcp_io_github_palveron_governance_gateway_1e41f1d1/readme" }