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.
npx agentmods add commands/areu01or00/perf-hints/perfgit clone --depth 1 https://github.com/areu01or00/perf-hintsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/areu01or00/perf-hints/perf)<a href="https://agentmods.dev/commands/areu01or00/perf-hints/perf"><img src="https://agentmods.dev/badge/commands/areu01or00/perf-hints/perf.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00016 | $0.01141 |
| Opus 5 | $0.00008 | $0.00571 |
| Sonnet 5 | $0.00003 | $0.00228 |
| Haiku 4.5 | $0.00002 | $0.00114 |
Grade A, and why
perf 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Performance Optimization
You are helping a developer optimize code for performance. Follow Jeff Dean and Sanjay Ghemawat's systematic approach: estimate first, measure, identify bottlenecks, then fix with validation.
Core Principles
- Estimate before optimizing: Back-of-envelope calculations tell you if optimization matters
- Quantify everything: "Faster" means nothing without numbers
- Fix the right thing: Profile shows where time goes, not what to fix
- Validate fixes: Optimizations that break correctness are worse than slow code
- Use TodoWrite: Track all progress throughout
Phase 1: Scoping
Goal: Understand what needs optimization
Target: $ARGUMENTS
Actions:
- Create todo list with all phases
- If scope unclear, ask user:
- What's slow? (specific endpoint, feature, operation)
- How slow is it now?
- What's the target?
- Any constraints? (can't change API, memory limits, etc.)
- Confirm understanding before proceeding
Phase 2: Analysis
Goal: Back-of-envelope estimation and bottleneck detection
Actions:
-
Launch 2-3 perf-analyzer agents in parallel, each focusing on:
- Different parts of the codebase (e.g., API layer, data layer, external calls)
- Or different optimization angles (I/O patterns, memory patterns, algorithmic complexity)
Example agent prompts:
- "Analyze the [endpoint/feature] for I/O bottlenecks - sequential calls, missing parallelization, client reuse"
- "Do back-of-envelope calculation for [operation] - estimate latency breakdown"
- "Find allocation patterns and memory inefficiencies in [module]"
-
Once agents return, read all files they identified
-
Present consolidated findings:
- Back-of-envelope latency table
- Bottlenecks ranked by impact
- Estimated total improvement possible
Phase 3: Clarifying Questions
Goal: Resolve ambiguities before implementing
CRITICAL: Do not skip this phase.
Actions:
- Based on analysis, identify decisions that need user input:
- Trade-offs (memory vs speed, complexity vs performance)
- Scope (fix everything vs highest ROI only)
- Constraints (can't change external APIs, must maintain compatibility)
- Present questions to user
- Wait for answers before proceeding
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.
- 4d ago First seen · 163 lines · 16 tokens per session scan A fcbf9b3768f9
perf is a command published in the GitHub repository areu01or00/perf-hints (4 stars, last pushed 8mo ago), licensed MIT. It adds 16 tokens to every session and 1,141 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.
Other commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.