Capacity Ledger Desk

Cost-first memory layer for MCP-capable agents that stores memories cheaply and recalls relevant slices under a hard token budget, logging receipts for every recall.

Enables governed Codex work sessions by planning bounded segments, maintaining capacity reserves, and creating checkpoint-based resume contracts so long-running jobs can pause and resume deterministically within quota limits.

Fleet-wide shared rate limiter for A2A + multi-MCP deployments, providing a shared counter with sliding window, concurrency grants, and signed enforcement attestations.

Enables AI agents and multi-agent frameworks to throttle outbound LLM API calls with multi-provider token-bucket and sliding-window limiting, plus predictive token-consumption budgeting to avoid TPM/RPM throttling. It runs deterministically over MCP with zero dependencies, returning rate-limit decisions and telemetry for any client such as Claude Desktop or Cursor.