Discover the toolingbehind modern intelligence
The operating system for AI infrastructure discovery. Search a live map of MCP servers, AI agents, LLM tools, automation systems, and developer infrastructure.
headroom
Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers. Library, proxy, MCP server.
11.8kstars
770forks
Python
agentaianthropicclaude-codecompressioncontext-engineeringcontext-windowcursorfastapilangchainllmmcp
Quality signals
ActiveNo testsNo CIThin docsNo licensePopular
AI summary
Auto-generated overview of what this skill does and when to reach for it.
Compresses AI agent outputs, logs, and files by 60-95% before they reach the LLM without changing answers.
Use this when
- Running daily coding agents and want to cut token costs
- Debugging with long logs or tool outputs
- Using multiple AI agents and need shared memory
- Working with RAG chunks that exceed LLM context limits
- Needing reversible compression to retrieve originals on demand
Don't use when
- Only using a single provider’s built-in compaction
- Working in a locked-down environment where local processes can’t run
Capabilities
- Compresses JSON, code, logs, and text with 6 algorithms
- Acts as a library, proxy, or MCP server
- Shares memory across agents like Claude, Codex, Cursor
- Reversible compression (CCR) stores originals locally
- Auto-detects content type and selects the right compressor
- Provides cross-agent memory and correction learning
InstallMCPRun via npx (Node)
Universal installANY TOOL
$ npx skillcity@latest chopratejas/headroomAuto-detects the skill or MCP server in this repo and installs it for Claude Code, Codex, Cursor or Cline.
Requires Node.js 18+ (ships with npx).
Or install manually
npx -y headroom
MCP Configuration
Add this to your claude_desktop_config.json
{
"mcpServers": {
"headroom": {
"command": "npx",
"args": [
"-y",
"headroom"
]
}
}
}Loading README…
Ask a simulated agent to run this skill. It loads the real SKILL.md and reasons over your task with NVIDIA.
headroom — sandboxAI sandbox
Pick a quick prompt below or type your own — watch the agent execute this skill step by step.
guest@skillcity:~$Try a prompt