agentleFS
Sign inStart free
gclmgclm/super-stack/skills/quality/performance-investigation/SKILL.md

performance-investigation

Skill0 starsChanged 7 months agoMITView on GitHub

Investigate latency, throughput, memory, and CPU issues with evidence-first profiling and bottleneck isolation before applying speculative optimizations.

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when the problem is slow response time, poor throughput, memory growth, CPU pressure, rendering jank, or unclear runtime …

    line 8
  2. Distinctive:

    Test one optimization hypothesis at a time.

    line 31
  3. Very specific:

    Read references/perf-evidence.md when you need more detailed evidence patterns.

    line 35

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Performance Investigation

Use this skill when the problem is slow response time, poor throughput, memory growth, CPU pressure, rendering jank, or unclear runtime bottlenecks.

Read First

  • current symptom and measured baseline
  • profiles, traces, benchmarks, or logs if available
  • the hottest path or user-visible slow flow
  • harness/state.md if it exists
  • harness/history.md if it exists
  • references/perf-evidence.md when you need a sharper measurement workflow

Goals

  • identify the dominant bottleneck before changing code
  • distinguish measurement from assumption
  • focus on the slowest meaningful path first
  • recommend evidence-backed fixes, not speculative tuning

Steps

  1. Restate the performance symptom with numbers if possible.
  2. Identify the measured hot path.
  3. Compare likely bottlenecks against actual evidence.
  4. Test one optimization hypothesis at a time.
  5. Recommend the smallest fix that materially improves the dominant bottleneck.
  6. Define how improvement should be re-measured.

Read references/perf-evidence.md when you need more detailed evidence patterns.

Output

Report:

  • symptom and baseline
  • evidence used
  • dominant bottleneck
  • recommended optimization
  • expected trade-offs
  • re-measurement plan

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free