MCP servers and skills: search, filter by where they run, and compare.
Context compression plugin for Claude Code. Automatically trims large tool output—JSON, YAML, stack traces, and logs—before it enters the context window.
Node.jsMIT★ 67No key
transparent TOON re-encoding for AI coding agent tool-call context
SkillAGPL-3.0★ 2Few starsNo key
Token optimization MCP server that reduces LLM context by stripping code comments and converting data formats to TOON
PythonMITFew starsNo key
MCP server giving AI coding agents offline, token-efficient persistent memory — recall decisions, patterns, and bugs across sessions with 68.5% fewer tokens than re-reading context. No LLM, no server;
Node.jsMIT★ 12No key
Toonkit image, video, and canvas plugin for Codex and Claude Code
SkillMIT★ 2Few starsNo key
🚀 Claude Code hook that automatically compresses JSON data by up to 60% using Toon format before sending to LLMs. Save tokens, reduce costs, same results.
SkillMIT★ 9No key
Lightweight MCP server for Linear with TOON-formatted responses for token-efficient LLM interactions
SkillMIT★ 1Few starsNo key
Smart JSON-to-TOON conversion with pragmatic auto-gating for LLM prompts
PythonMIT★ 2Few starsNo key
GCF Python implementation. 100% LLM comprehension on every frontier model. 50-92% fewer tokens than JSON. 43B+ round-trips verified. Zero dependencies.
PythonApache-2.0★ 5No key
GCF TypeScript implementation. 100% LLM comprehension on every frontier model. 50-92% fewer tokens than JSON. 43B+ round-trips verified. Zero dependencies.
Node.jsApache-2.0★ 4Few starsNo key
GCF Swift implementation. 100% LLM comprehension on every frontier model. 50-92% fewer tokens than JSON. 43B+ round-trips verified. Zero dependencies.
SkillApache-2.0★ 2Few starsNo key
TOON-Native Auto-Embedding & RAG Toolkit for MariaDB | Judges' Guide & live demo: https://mariadb-hackathon-my-2026.github.io/seamless-rag/
SkillApache-2.0★ 3Few starsNo key
A Claude Code skill that performs static analysis on JavaScript files to find API endpoints, URLs, secrets, emails, and file references. Supports TOON format output (50% smaller than JSON) optimized f
SkillMIT★ 9No key
Production pipeline and Claude Code skill for Vocaloid-style hand-drawn (手書き) lyric music videos: deterministic Canvas2D/WebGL2 animation engine, JIZURA lyric motion, AI illustration, Blender toon 3D,
SkillMIT★ 213No key
⚜️ An MCP server for context compaction and recycling in Claude Code
Node.jsMIT★ 11No key
No-code MCP server generator from API specs (OpenAPI, GraphQL, protobuf). TOON compression for fewer tokens. AI agent for large response pruning and prompt injection detection.
PythonMITFew starsNo key
AXI-compliant glab CLI wrapper for GitLab - token-efficient TOON output for agents. Works with gitlab.com and self-hosted.
SkillMIT★ 8No key
AST CLI for AI agents that parses 16 languages with tree-sitter and reports symbols, references, diff impact, API changes, and dead code as JSON or TOON
SkillMIT★ 2Few starsNo key
An agentic memory database that cuts session tokens by 82–99%. One portable SQLite file — your agent's memory, anywhere.
GoMIT★ 129No key
Agent skills for building a hand-painted, Ghibli-style 3D world in three.js / WebGL — water, toon foliage, sky, night, weather, creatures, interaction, tours, Blender → GLB pipeline
SkillMIT★ 3Few starsNo key
Shrinks your AI agent's tool output so it burns fewer tokens, without changing what the agent reads. Lossless by default, no decode step.
PythonMIT★ 2Few starsNo key
Agent-first Forgejo CLI with compact TOON output, token-safe dry runs, an installable agent skill, and a Codex plugin.
SkillMITFew starsNo key
Who stands behind it
