context-packing

TokenPack packs long documents, codebases, PDFs, and folders into compact, evidence-dense LLM context using local embeddings, evidence scoring, and budget-aware selection.

📚 Batch file reads for Pi via read_many with adaptive packing, safe heredoc framing, and output-budget awareness

Pack local folders into single LLM-friendly context files for ChatGPT, Claude, Gemini, and automation.

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.

Pack your entire codebase into a single clean Markdown file. Ready for LLMs and AI-assisted code review.

Preprint paper package — Context Forge: A Lightweight Method for Diversity-Aware Context Packing and Prompt-Injection-Aware Retrieval (Research Square pending)

Safety-first, local-first CLI that gives AI coding agents and developers one guarded interface to a repository: scoped search, LLM-ready context packing, edits with rollback, test selection, Git inspection, and evidence-based verification.

Enable efficient batch reading of multiple files in Pi to reduce calls and streamline file inspection with a single tool.