llm-finops
High-assurance in-memory Tree-Sitter AST context firewall and pruning MCP server for coding agents (-72.4% token mass).
Python SDK for pre-call AI budget controls and customer-, feature-, and workflow-level cost attribution.
Tamper-evident, stranger-verifiable receipts for LLM cost-savings — anyone can recompute your caching/routing savings math offline, no trust in your dashboard required. Pure stdlib, zero-dependency.
OTel-native typed primitives for LLM cost attribution and telemetry — published on PyPI.
