A pure-Python reachability and skeletonization layer that decides what a coding agent actually sees before a request reaches the API.
Deterministic context and policy compiler for supported AI coding agents.
Deterministic state engine for managing conversation state and constraints in LLM applications.
Context compiler for AI agents — build compact, evidence-aware Context Capsules from memory, state, files and tools over MCP. Compile context, don't concatenate it.
Semantic context compiler for LLM coding agents — packs the smallest useful codebase context for any task. Zero dependencies, deterministic, token-efficient.
Trust-domain aware context compiler and agent-facing frontend for CIC agent context (cic-mcp-* family)
Repo-native context compiler for coding agents. Compiles rules, handbook truth, and ASTs into SQLite where the graph is the API — query SQL for instant code-graph projections, call flows, and rapid pinpointing beyond raw grep.
Portable AI Project Context Compiler with incremental change journaling, deterministic AST graphs, and token-budgeted handoffs (ctxa).
TypeScript port of the Context Compiler core
Review-gated context compiler — turn long-term content into multi-runtime agent context (CLAUDE.md, AGENTS.md, …) with PR/approve gates and before/after evaluation.
A context compiler for AI coding agents: state intent in English, get governed context (CLAUDE.md, agent defs, checks, provenance) for Claude Code and Codex.
TypeScript directive drafting and acquisition-layer utilities for Context Compiler.
Draft candidate Context Compiler directives from natural-language input.
To associate your repository with the context-compiler topic, visit your repo's landing page and select "manage topics."