Kompact — AI skill for Claude Code
LLM context compression proxy — 40-70% token savings, zero code changes.
How to install Kompact
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open npow/kompact and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Kompact does
LLM context compression proxy — 40-70% token savings, zero code changes.
Alternatives in AI
- Opencodex — Universal provider proxy for OpenAI Codex & Claude Code — use any LLM (Claude, Gemini, Grok, DeepSeek, Ollama… 12.3k ★
- WindsurfAPI — Turn Windsurf / Devin Desktop's 100+ AI models (Claude, GPT, Gemini, DeepSeek, Kimi, GLM, SWE) into OpenAI-, A 3k ★
- Paritok 4b V1 — Non-destructive compression gateway for AI coding agents 1.4k ★
README
Kompact
[](https://github.com/npow/kompact/actions/workflows/ci.yml) [](https://pypi.org/project/kompact/) [](https://www.python.org/downloads/) [](https://mintlify.com/npow/kompact)
Context compression proxy for LLM agents. Sits between your agent and the LLM provider, compresses context on the fly, and cuts your token bill 40-70% — with zero code changes.
Save real money
For a team running 1,000 agentic requests/day with ~10K token contexts:
| Model | Without Kompact | With Kompact | Monthly Savings |
|---|---|---|---|
| Sonnet ($3/M) | $900/mo | $405/mo | $495/mo |
| Opus ($15/M) | $4,500/mo | $2,025/mo | $2,475/mo |
| GPT-4o ($2.50/M) | $750/mo | $338/mo | $412/mo |
Savings scale linearly. 10K requests/day = 10x the numbers above.
Get started in 30 seconds
pip install kompact # or: uv add kompact
kompact proxy --port 7878export ANTHROPIC_BASE_URL=http://localhost:7878
# That's it. Your agent now uses fewer tokens.No SDK changes. No prompt rewriting. Just point your base URL at the proxy.
Quality stays intact
Evaluated on [BFCL](https://gorilla.cs.berkeley.edu/) (1,431 real API schemas) — the standard benchmark for tool-calling agents. End-to-end through Claude, scored with [context-bench](https://pypi.org/project/context-bench/).
Quality impact vs no compression (closer to 0% = better):
| Model | Kompact | Headroom | LLMLingua-2 |
|---|---|---|---|
| Haiku | -2.6% | -3.0% | -23.4% |
| Sonnet | -3.9% | -3.5% | -20.6% |
| Opus | -0.5% | -0.5% | -27.3% |
Kompact and Headroom bo
Related Skills
Claude Codex Cursor Token Saver
AI Coding Token Saver for Windows — optimize prompts and repository context before sending it to Claude, Codex
Cost Optimizer
git clone https://github.com/fullstackcrew-alpha/skill-cost-optimizer Save 60-80% on AI token costs with smart
.Git Z2ey1tgs8e
AI Coding Token Saver for Windows — optimize prompts and repository context before sending it to Claude, Codex
Lm Resizer
Rust-native context compression for Claude Code, Codex & MCP agents: filters & compresses noisy tool output (t
Distil
Measure what context compression actually costs an LLM agent: where a session's tokens go, and whether a rewri
Loongsuite JS
OpenTelemetry instrumentation plugins for JavaScript-based AI coding agents. Collect traces, tool calls, and L
Related Agents
Codemap Writer
Generate or refresh docs/CODEMAPS/ for the current repo. Use when the user runs /update-codemaps, when context
Cold Reviewer
Zero-context code reviewer. Reviews changes cold - no spec, no project knowledge, no domain context. Catches w
Context Master
Use when context exceeds ~85% of the window, or during WF closeout to extract durable knowledge before compres
