# Context Bench

> Plugin description: Self-learning Claude Code plugin that automatically detects what project you're working on and injects the right file context into every session. No manual configuration needed — it learns from your usage patterns, builds confidence scores, and removes stale topics automatically. Uses three hooks: UserPromptSubmit (match & inject), PostToolUse (track changes), SessionEnd (learn & decay). Zero dependencies, Python 3.9+ stdlib only.

## Facts
- Page: https://tashan.sh/capability/plugin-nessos666-context-bench-context-bench
- tashan id: plugin:nessos666/context-bench/context-bench
- Source: https://github.com/nessos666/context-bench
- Type: plugin
- Category: devtools
- tashan score: 36.0 / 100
- Adoption: 13.0
- Upkeep: 69.0
- Freshness: 72.0
- Evidence coverage: 84% of the inputs this score can use
- Health: active
- Instruction depth: not yet graded
- GitHub stars: 2
- License: MIT
- Official: no

## Install

```sh
/plugin marketplace add anthropics/claude-plugins-community
/plugin install context-bench@claude-community
```

## Security audit
Not scanned. We audit npm-published capabilities; this one has no npm package we can resolve, or has not reached the queue. This is not a clean bill of health.

---
Measured 2026-09-13 by tashan (https://tashan.sh) from public evidence. Scorer s5.
