#context-window

Evidence-based pruning for your AI-agent stack — scan, report, and safely prune unused skills/MCP servers/agents using real transcript evidence

MCP-native code retrieval for AI agents — 84-88% fewer read tokens, BM25F + semantic search, AST chunks, session dedup

Local-first agentic memory lifecycle manager for AI coding IDEs

The Best Srcpack Alternatives 2026: Optimize Code for LLMs Faster

Community catalog of LLM model specs — pricing, context windows, capabilities, and AI-CLI compatibility. Auto-synced from LiteLLM with objective corrections.

Local-first context cleaner for AI agents.

Local‑only Query Risk & Scope Analyzer for AI coding prompts — estimate risk, scope, likely files, and token pressure before sending queries to agents like Cursor, Copilot, or Claude.

Stop Claude Code from silently dying mid-task — injects quota/context signals into the model so it converges gracefully

A four-line statusline for Claude Code: usable context, burn rate, rate-limit pace projection, cross-session pool.

Automatic session handoff for coding agents — never hit a context limit again. Claude Code & Codex CLI.

OpenAI-compatible context gateway & proxy for local LLMs (LM Studio, Ollama, etc.). Enforces strict token budgets, deterministic archival, and safe tool-calling for AI agents.

In-flight conversational context compressor and stale-turn pruner to eliminate context rot.