# Perf · agent-sh

> Structured performance investigation with baselines, profiling, and evidence-backed decisions — run by AI agents inside your editor. /perf enforces a rigorous 10-phase methodology derived from recorded real investigation sessions: establish a baseline, identify the breaking point, generate hypotheses, profile, run controlled experiments, measure, document. No random micro-optimizations; every change is baselined, isolated, measured, and either kept or reverted with evidence. Designed to keep t

## Facts
- Page: https://tashan.sh/capability/plugin-agent-sh-perf-perf
- tashan id: plugin:agent-sh/perf/perf
- Source: https://github.com/agent-sh/perf
- Type: plugin
- Category: other
- tashan score: 40.0 / 100
- Adoption: 13.0
- Upkeep: 74.0
- Freshness: 82.0
- Evidence coverage: 84% of the inputs this score can use
- Health: active
- Instruction depth: not yet graded
- GitHub stars: 2
- Official: no

## Install

```sh
/plugin marketplace add anthropics/claude-plugins-community
/plugin install perf@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.