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.

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

Free local CLI that estimates, hard-caps, and losslessly compresses the cost of AI coding agents. Delta-encodes re-read files (37.9% proven on a real OpenAI call). MIT, 100% local.

Self-hosted spend firewall and gateway for LLM ( OpenAI / Anthropic / Gemini ). Hard per-user & per-project budget caps that block runaway costs before the API call, plus cost-per-customer tracking, semantic caching, and failover. One line of code, single Go binary.

⚡ ZeroToken Skill — 让 Agent 用最少的 token 做最准的事。提示词纪律规范：压缩无效上下文、无效解释、无效工具调用、无效输出，准确性不降。

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

Coding agents forget your repo. mcp-brain is the missing memory layer — repo-aware, team-aware, lifecycle-aware. 63% Hit@10, zero LLM cost. Works with any MCP client.

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

Don't go into production without these - 3 auto-triggering Claude Code skills for cost and drift prevention. 6 months of practitioner notes.

A drop-in SKILL that forces AI coding agents (Claude Code, Codex, Cursor, Cline, Roo, Windsurf, Copilot, Augment, Aider, …) to deliver exactly what was asked — minimum diff, zero unsolicited files, terse output by default.

Open source AI cost tracking. Know exactly what your AI costs — per feature, per user, per project.

TypeScript SDK that adds cost limits, token/call budgets, timeouts, and circuit breakers to AI agent/LLM workflows, with adapters for OpenAI, Anthropic, Vercel AI, and observability/reporting support.

Zero-dependency context-window packer for LLM chat: fit a conversation into a token budget (middle-out, drop-oldest, priority, pinning).

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

CLI for building, resolving, and inspecting context caches

Build optimal LLM context windows. Priority-based assembly, token budgeting, smart truncation (HEAD/TAIL/MIDDLE). Zero mandatory dependencies. Drop into any RAG pipeline in 3 lines.

Context window selection for LLM coding tools. Treats context as a constrained optimization problem, not retrieval.

Context engineering toolkit for LLMs — pack, cache, debug, red-team, and orchestrate context windows. Council of Experts, adversarial testing, immune system, context compiler, drift detection, multi-agent entanglement. TypeScript + Python.

ContextOps toolkit for production AI agents. Compile, govern, scan, visualize & optimize LLM context before every model call — token budgets, policy governance, PII/secret scanning, Context Bill of Materials, diffing, Context MRI, MCP budgeting. Framework-agnostic. Local-first. Deterministic.

A pre-flight budget gate for Claude Code multi-agent workflows. Gates subagent spawns against a self-managed token ledger. Works on Claude Pro/Max subscription.