Token Reducer

RAG context-compression pipeline for agents — its CLI script isn't in the skill folder.

Works with setup

Test report

Verdict
Works with setup
Score
6.8/10
Tested
Jul 14, 2026
Environment
Claude Code 2.x (agent harness)
Upstream re-checked
Aug 10, 2026 · da4a47a

The CLI it tells you to run — `${CLAUDE_PLUGIN_ROOT}/scripts/context_pipeline.py` — lives at the repo root, not inside the skill folder the README has you copy, so a bare skill install 404s on its own primary command.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

  • Installs cleanly 3/5
  • Triggers reliably 5/5
  • Output vs. baseline 7/10
  • Docs & honesty 2/5

What Token Reducer does

Runs a local BM25 + embeddings retrieval pipeline to chunk, rerank, and compress large code or doc corpora into a compact, citation-rich context packet before implementation work. Triggers on phrases like 'reduce tokens', 'context is too long', or 'retrieve top chunks'. The skill folder alone can't run its own advertised CLI command — the referenced script and settings.json live in the parent plugin repo root, not inside the directory the SKILL.md tells you to copy.

How to install Token Reducer

git clone https://github.com/Madhan230205/token-reducer
cd token-reducer
mkdir -p ~/.claude/skills
cp -r skills/token-reducer ~/.claude/skills/token-reducer

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Token Reducer

  • /token-reducer RAG context-compression pipeline for agents — its CLI script isn't in the skill folder.

It also activates on plain-language prompts like these:

  • This context is too long, compress it before we continue
  • Reduce tokens in this huge doc before we implement the feature
  • Retrieve just the top relevant chunks from this large codebase

Frequently asked questions

Is the Token Reducer skill free?
Yes. The skill itself is free from Madhan230205/token-reducer. SkillProof publishes the install command and an independent test verdict at no cost.
Does Token Reducer work with Claude Code?
We tested it with Claude Code 2.x (agent harness) on Jul 14, 2026. Verdict: Works with setup. The CLI it tells you to run — `${CLAUDE_PLUGIN_ROOT}/scripts/context_pipeline.py` — lives at the repo root, not inside the skill folder the README has you copy, so a bare skill install 404s on its own primary command.
What is the Token Reducer SkillProof Score?
6.8/10 — installs cleanly 3/5, triggers reliably 5/5, output vs. baseline 7/10, docs & honesty 2/5.
How do I install Token Reducer?
Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Token Reducer with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.