{
  "markdown": "# klanex MCP server\n\n[![npm](https://img.shields.io/npm/v/klanex-mcp)](https://www.npmjs.com/package/klanex-mcp)\n\nReliable execution for agent tool calls, as an [MCP](https://modelcontextprotocol.io) server.\n\n[klanex](https://klanexai.com) is an async execution layer between an agent's tool-use intents and real third-party APIs: it schema-gates hallucinated payloads before anything runs, absorbs rate limits and outages with retries/backoff/circuit breakers, encrypts credentials, keeps an audit trail, and reports back via signed webhooks. With this MCP server, an agent submits a call once — klanex makes sure it lands.\n\nFailures come back written *for the model*: every error carries an `llm_hint` telling the agent exactly what to fix (or that klanex is already handling it), so agents self-correct instead of guessing.\n\n## Connect (hosted, recommended)\n\nklanex's MCP server is hosted — no install needed if your client supports HTTP transports:\n\n| Endpoint | Environment | Keys |\n|---|---|---|\n| `https://api.klanexai.com/mcp` | production | `klx_live_…` |\n| `https://api.sandbox.klanexai.com/mcp` | sandbox (free to try) | `klx_test_…` |\n\nGet an API key at [klanexai.com](https://klanexai.com) — sandbox keys are free.\n\n**Claude Code**\n\n```bash\nclaude mcp add --transport http klanex https://api.klanexai.com/mcp \\\n  --header \"X-API-Key: klx_live_...\"\n```\n\n**Cursor / Windsurf / VS Code** (`mcp.json`)\n\n```json\n{\n  \"mcpServers\": {\n    \"klanex\": {\n      \"url\": \"https://api.klanexai.com/mcp\",\n      \"headers\": { \"X-API-Key\": \"klx_live_...\" }\n    }\n  }\n}\n```\n\n`Authorization: Bearer klx_...` works too, for clients that can only set that header.\n\n## Connect (stdio shim)\n\nFor clients that only speak stdio (e.g. Claude Desktop), this package proxies stdio to the hosted endpoint:\n\n```json\n{\n  \"mcpServers\": {\n    \"klanex\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"klanex-mcp\"],\n      \"env\": { \"KLANEX_API_KEY\": \"klx_live_...\" }\n    }\n  }\n}\n```\n\nThe shim routes by key prefix — `klx_test_…` keys go to the sandbox automatically. Set `KLANEX_MCP_URL` to override.\n\n## Tools\n\n| Tool | What it does |\n|---|---|\n| `execute` | Submit an HTTP call for reliable async execution: optional JSON Schema gate, retries/backoff/circuit breakers, idempotency keys, human approval gates, encrypted credentials. `wait_seconds` blocks up to 55s for the terminal result. |\n| `get_execution` | Current status, attempts, target response, or classified error for an execution. |\n| `list_executions` | The account's executions, filterable by status/time, paginated. |\n| `replay_execution` | Re-run a terminal execution byte-exact — outage recovery without re-prompting the LLM. |\n| `get_usage` | Plan and current-month usage/quota. |\n| `list_connections` | Stored credential connections (token vault) to reference via `connection_id` — so secrets never pass through model context. |\n\n## Why this beats calling APIs directly\n\n- **Hallucinated payloads are rejected before they execute** — pass `payload_schema` and mismatches return synchronously with a correction hint.\n- **Transient failures are not the agent's problem** — 429s, timeouts, and 5xxs are retried with backoff behind a circuit breaker; the agent does nothing.\n- **A network blip can never double-execute a refund** — `idempotency_key` makes submits safe to retry.\n- **Destructive actions can wait for a human** — `requires_approval` pauses execution for an approve/reject decision (API, dashboard, or Slack buttons).\n- **Credentials stay out of model context** — store them once as a connection; klanex injects them at execution time.\n\nFull API docs: [api.klanexai.com/docs](https://api.klanexai.com/docs)\n\n## License\n\nMIT\n",
  "bytes": 3679,
  "sha": "2c55128951c1511a0e632ebd43fd516dee727cf0e64a31ac6ff2aa10e760f513",
  "repo_slug": "chrassy/klanex-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_chrassy_klanex_30c39103/readme"
}