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 skills/stankye/profiler-mcp/optimize-loopnpx skills add Stankye/profiler-mcp --skill optimize-loopgit clone --depth 1 https://github.com/Stankye/profiler-mcpWhat 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 | $0.00052 | $0.00586 |
| Opus 5 | $0.00026 | $0.00293 |
| Sonnet 5 | $0.00010 | $0.00117 |
| Haiku 4.5 | $0.00005 | $0.00059 |
Grade A, and why
optimize-loop 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.
How it starts
The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Optimize loop
The loop that turns a profiler into an optimizer. The compare tool's verdict is the gate: no change is kept on vibes.
Protocol
- Baseline. Collect with fixed, repeatable workload parameters
(
command=[app, "--seconds", "5"]style; same machine, same load). Save theresult_id— say it out loud in your response so it survives context loss. Run*_report_hotspotsand pick ONE target: the top function you can actually change. - Hypothesize before editing. From the hotspot's signature (high
percent+extrametrics like CPI/cache-miss where present) state what class of fix applies: algorithm, memory layout, branch shape, parallelism. One hypothesis, one change. - Change. Smallest edit that tests the hypothesis. Rebuild with identical flags.
- Verify correctness first. Run the program's own tests/output check. A fast wrong program is a failure — never skip this step.
- Re-profile with the exact same collection parameters → new
result_id. - Compare.
*_compare(baseline_id, candidate_id, threshold_pct=5, fail_on_regression=true).- Tool errors with the verdict reason → the change regressed: revert, return to 2 with the next hypothesis.
verdict.passedand total improved → keep the change; the candidate becomes the new baseline id.passedbut total within threshold → treat as neutral; keep only if it also improves clarity, else revert.
- Repeat until the target function is off the top of the profile, the improvement goal is met, or two consecutive hypotheses fail (then stop and report honestly).
Rules
- Never compare results from different modes (
realvsmock) or different machines — the tools note mode mismatches; treat them as invalid comparisons. - Sampling noise is real: below ~5% delta on short runs, lengthen the workload before believing a win.
- Keep a running table in your response: hypothesis → delta% → kept/reverted. That table is the deliverable, alongside the final diff.
- In mock mode this loop exercises the machinery (the mock's numbers won't respond to your code edits — weights are fixed). Use mock mode to test the loop itself, real mode to optimize actual programs.
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.
- 2d ago First seen · 45 lines · 52 tokens per session scan A 8a1b0203542e
optimize-loop is a skill published in the GitHub repository Stankye/profiler-mcp (0 stars, last pushed 28d ago), licensed MIT. It adds 52 tokens to every session and 586 once invoked, about $0.0003 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 skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
brainstorming
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
auto-perf-optimize
Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.
chat-perf
Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
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…