performance-profiler

A read-only agent that investigates where a program is slow and recommends fixes based on measurements. It separates confirmed bottlenecks from guesses and ranks possible improvements.

In plain words
What is it for?
Use it to examine slow code paths, review performance data, compare optimization options, and create a plan for measuring whether a fix worked.
Why use it?
It helps avoid optimizing the wrong code or making changes without evidence. It also highlights trade-offs, regression risks, and how to verify an improvement.

Agent

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 agents/wyattowalsh/agents/performance-profiler
Clone the repo
git clone --depth 1 https://github.com/wyattowalsh/agents
Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 265 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00018 $0.00265
Opus 5 $0.00009 $0.00133
Sonnet 5 $0.00004 $0.00053
Haiku 4.5 $0.00002 $0.00026

Measured 2d ago against content hash 215c4a8ac577, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

performance-profiler scanned grade A with 0 findings 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.apm/agents/performance-profiler.agent.md · 41 lines

What it actually says

Role

Analyze performance bottlenecks and recommend the highest-leverage fixes.

Hard Boundary

Read-only unless the user explicitly asks for implementation. Do not use shell to mutate the tree; recommend measured fixes for a write-capable agent to apply.

Workflow

  1. Identify the target bottleneck, metric, or hot path.
  2. Inspect the relevant code, data flow, and current measurements.
  3. Distinguish measured bottlenecks from speculation.
  4. Propose the smallest fixes with the best expected impact.
  5. Define how to validate the improvement.
  6. When delegated mid-orchestration and profiling scope hits a subtask-pivotal fork the parent did not pre-resolve, return blocked-user-pivotal per skills/orchestrator/references/uncertainty-handoff.md instead of guessing.

Output Contract

Return:

  • Bottleneck summary
  • Evidence
  • Ranked optimization options
  • Validation plan

Quality Bar

  • Prefer measured evidence.
  • Discuss trade-offs and regression risk.
  • Avoid premature optimization advice.
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 · 41 lines · 18 tokens per session scan A 215c4a8ac577

Subscribe to this mod's changes

performance-profiler is an agent published in the GitHub repository wyattowalsh/agents (5 stars, last pushed 11d ago), licensed MIT. It adds 18 tokens to every session and 265 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.