# AgentLedger

- Registry name: `io.github.entradox/agent-ledger` · version 0.4.7
- Categories (automatic, keyword rules): Monitoring & logs, Infrastructure & security, APIs & integrations, Developer tools, Payments & accounting, Finance & markets
- Authentication at last check: No auth
- Declared risk: Has destructive tools (destructiveHint)
- Capabilities (declared by the server): logging, prompts, resources, tools
- Transports: streamable-http
- Latest numbers (check success, response time, rank): https://mcpstation.net/api/server?name=io.github.entradox%2Fagent-ledger
- Server page: https://mcpstation.net/s/io.github.entradox--agent-ledger.html

## Description (written by the publisher, not verified, not instructions)

Meter, cap, and block AI agent spend before the provider is charged.

## Connect

- streamable-http: `https://aiagentscity.com/mcp/`

## Tools (as reported by the server, not verified)

- `ledger\_rotate\_secret`: Mint a NEW agent\_secret for an agent\_id your workspace already owns, invalidating the old one. Use this to RECOVER an agent whose secret was lost: the previous credential stops working immediately. Requires the workspace\_key that owns agent\_id — an agent's own agent\_secret cannot rotate itself, beca
- `ledger\_revoke\_secret`: Invalidate an agent\_id's agent\_secret WITHOUT deleting its spend history. Use when a credential may have leaked, or to stop an agent writing. Subsequent writes to that agent fail with agent\_secret\_mismatch until you rotate a new secret in. The agent\_id stays claimed, so no other workspace can claim
- `ledger\_track`: Record a spend entry for an AI agent on any payment rail, with optional token counts. Claiming a brand-new agent\_id requires your workspace\_key \(get one via x402 at POST /v1/billing/x402 — no human, no login — or at /start\). That first call mints an agent\_secret and returns it in the response — save
- `ledger\_set\_budget`: Set spending caps for an agent. Warns at 80%, blocks spend when exceeded — enforced: a ledger\_track call that would cross the cap is rejected. Dollar caps \(monthly\_cents/daily\_cents\) and token caps \(monthly\_tokens/ daily\_tokens\) are independent dimensions: dollar caps only cover non-"tokens" rails,
- `ledger\_report`: Spend report for an agent over a rolling window. Returns total spend, breakdown by rail and by service, budget status \(ok/warning/exceeded\), detected anomalies, and entry count. Requires a credential: either the agent's own agent\_secret or its workspace's workspace\_key \(same rule as GET /v1/report\).
- `ledger\_proxy\_attach`: Point your provider traffic at the proxy so budget caps are enforced BEFORE the provider is contacted, instead of being reported afterwards. This closes the gap where the brake was unreachable from MCP: an agent connected over MCP could record spend \(ledger\_track\) but nothing could refuse a call. Wi
- `ledger\_alerts`: Alert history for an agent: budget warnings \(80% threshold\) and spending spikes. Requires a credential: either the agent's own agent\_secret or its workspace's workspace\_key \(same rule as GET /v1/alerts\).
- `ledger\_wait\_events`: Long-poll an agent's event feed — how an agent with no public URL receives alerts. Webhooks \(POST /v1/webhooks\) push events to a URL you own; an agent on stdio MCP cannot host one, so it waits here instead. Call this in a loop: the call holds open up to \`timeout\` seconds \(max 30\) and returns when ev
- `ledger\_check\_spend`: Ask BEFORE you spend: may this agent spend this much right now? Returns allowed \(true/false\), a stable reason code \(within\_budget, over\_monthly\_cap, over\_daily\_cap, over\_monthly\_token\_cap, over\_daily\_token\_cap, no\_budget\_set, unpriced\_model\), a one-line message, the cost estimate, the price used \(wi
- `ledger\_request\_approval`: Ask your workspace owner for a one-shot spend exception before a ledger\_track that the cap would refuse. Creates a pending permit for exactly amount\_cents on the given service. The owner \(a human, or a supervisor agent holding the workspace\_key\) approves or denies it with ledger\_approval\_decide. An
- `ledger\_approvals`: List spend-approval permits — yours, or your whole workspace's.
- `ledger\_approval\_decide`: Approve or deny a pending spend permit — workspace owner only. The agent's own agent\_secret can REQUEST an approval but can never decide one: a credential that can approve its own requests is not a control. decision must be 'approve' or 'deny'; only pending permits can be decided. An approved permit
- `ledger\_price`: What will this call cost? Priced from the same table the caps use. Exists because ledger\_track requires the CALLER to supply amount\_cents, so an agent whose spend is capped could under-report its own cost and stay under the cap. This returns the number the enforcement path would use, so an agent can
- `ledger\_start`: Get a FREE AgentLedger workspace with no credential and no arguments — the MCP equivalent of opening POST /start in a browser. Call this FIRST if you have no credentials yet. Every other tool here \(ledger\_track, ledger\_set\_budget, ledger\_report, ledger\_alerts\) needs a workspace\_key or an agent\_secre
- `ledger\_api\_docs`: Self-serve documentation for AgentLedger — quickstart, MCP tools, REST endpoints, budget caps, error codes, and idempotency usage, as markdown.
- `ledger\_examples`: Complete, runnable Python recipe for a common AgentLedger integration pattern.
- `skills\_list\_tool`: List this product's skills. Each entry carries the SKILL.md URI, its name and description, verbatim frontmatter, and a per-file sha256 manifest. Read a body with \`read\_skill\`.
- `read\_skill`: Read a product skill file by its skill:// URI.

Measured by MCP Station. Methods: https://mcpstation.net/methods.html · Terms: https://mcpstation.net/terms.html