Open-source edge engine to control API request budgets and enforce fair usage.

Runtime containment kernel for LLM agents. Enforces budget, step, retry, and circuit-breaker limits before the model call.

AI agent harness engineering tool — visualize CLAUDE.md, AGENTS.md, skills, and agent config structure

🚀 Optimize AI context retrieval with OrionGraphDB, a powerful engine that respects token budgets and delivers diverse, relevant information seamlessly.

Core library: scoring, selection, and caching for the Context Engine

Cross-agent skill quality gate for SKILL.md files. Validates frontmatter, scores description discoverability, checks file references, enforces three-tier token budgets, and flags compatibility issues across Claude Code, VS Code/Copilot, Codex, and Cursor.

Governance layer for runtime budget, policy, and trade-off control in AI systems.

Token budget manager for LLM apps — track usage, enforce limits, estimate costs per user/session. Zero dependencies. OpenAI, Anthropic, Groq, Ollama.

A context database for AI agents. Multi-channel retrieval (semantic, lexical, structural) with MMR selection and token budget management. Built in Rust. Apache 2.0.

CLI for building, resolving, and inspecting context caches

Open-source platform for deterministic, token-aware context selection for AI agents and LLMs