profiling

A method for investigating software performance problems by collecting measurements before reading code or changing it. It covers issues such as slow queries, high latency, CPU spikes, memory growth, and lower throughput.

In plain words
What is it for?
Use it to investigate slow requests or queries, memory or CPU problems, throughput regressions, and performance changes after a deployment.
Why use it?
It reduces the risk of fixing an assumed cause instead of the measured one. The required order—profiling tools, logs, then source code—helps connect symptoms to evidence.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/cookys/autopilot/profiling
Any agent
npx skills add cookys/autopilot --skill profiling
Clone the repo
git clone --depth 1 https://github.com/cookys/autopilot

Made for: Claude Code, Codex.

Per session 91 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,150 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00091 $0.01150
Opus 5 $0.00046 $0.00575
Sonnet 5 $0.00018 $0.00230
Haiku 4.5 $0.00009 $0.00115

Measured 2d ago against content hash aa28f8017a80, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

profiling scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

!`cat .claude/profiling-config.md 2>/dev/null || echo "_No config — using generic tool selection below._"`
platforms/codex/plugin/skills/profiling/SKILL.md · 101 lines

How it starts

The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Evidence-First Profiling

Coexistence with Superpowers

This skill is the only methodology entry point for performance investigation in the autopilot + superpowers ecosystem. superpowers does not ship a dedicated profiling skill — CHANGELOG v2.0 acknowledged this explicitly. Whether or not you have superpowers installed, this is the primary skill for performance work.

Use this skill for: performance-only investigation (slow queries, high memory, CPU spikes, latency). For correctness debugging (crashes, logic errors), use autopilot:debug (or superpowers:systematic-debugging if installed).

Project Config (auto-injected)

!cat .claude/profiling-config.md 2>/dev/null || echo "_No config — using generic tool selection below._"

The Rule

Mandatory order — do not skip steps:

1. Collect evidence with tools (profiler, tracer, slow query log)
2. Analyze logs — correlate with tool results to locate the problem
3. Read source code — only after 1+2 provide evidence pointing to specific code

Prohibited: reading code to guess causes, modifying code based on "intuition", concluding without data.

Metric-honesty rule: an LLM reading static source cannot measure a real-world number (LCP, latency, throughput, memory) — it can only reason about likely causes. Label every such finding "potential impact", never as a measurement. A figure that didn't come from a tool run is a hypothesis, not a result; presenting it as measured is fabrication. Field data and lab/synthetic data are not interchangeable — don't quote one as the other. (This is the "verify by artifacts, never self-report" axiom applied to the one place an LLM is most tempted to invent a number.)

Tool Selection Guide

Pick the right tool for the symptom:

Symptom First Tool Why
CPU 100% or high load CPU profiler (gperftools, py-spy, node --prof) Shows which functions consume CPU time
Memory growing over time Heap profiler (gperftools, tracemalloc, heapdump) Tracks allocation sites and sizes
Slow requests / high latency Slow query log or APM Most latency comes from DB or external I/O
Crash (SIGSEGV, SIGABRT) Debugger (GDB, lldb) Get backtrace immediately
Unknown I/O bottleneck strace / dtrace Shows system call timing and blocking points
Need flame graph perf / async-profiler Hardware counters + call stacks

Read the full file on GitHub · 101 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 101 lines · 91 tokens per session scan B aa28f8017a80

Subscribe to this mod's changes

profiling is a skill published in the GitHub repository cookys/autopilot (11 stars, last pushed 2d ago), licensed MIT. It adds 91 tokens to every session and 1,150 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens