Borrowing it
Nothing to install: this file belongs to maguro/pbf. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/maguro/pbf/master/.claude/skills/go-benchmarks/SKILL.mdgit clone --depth 1 https://github.com/maguro/pbfWrote 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/skills/maguro/pbf/go-benchmarks)<a href="https://agentmods.dev/skills/maguro/pbf/go-benchmarks"><img src="https://agentmods.dev/badge/skills/maguro/pbf/go-benchmarks/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/maguro/pbf/go-benchmarks"><img src="https://agentmods.dev/badge/skills/maguro/pbf/go-benchmarks.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00057 | $0.00515 |
| Opus 5.5 | $0.00023 | $0.00206 |
| Sonnet 5.5 | $0.00011 | $0.00103 |
| Haiku 4.5 | $0.00006 | $0.00052 |
Grade A, and why
go-benchmarks 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 5d 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.
What it actually says
Measuring a change on a hot path
AGENTS.md lists the hot paths of this repo and the commands that measure them.
What gets measured
A change that can alter the work done on a hot path is measured before and after the change, on the same machine in the same session. This includes code on the path, default option values on the path, and updates to dependencies on the path and to the Go version.
When unsure whether a change qualifies, measure it. A change that does not alter the work done on the success path, such as an error message, error wrapping, a comment, or a rename, is not measured.
How to measure
Alternate the base runs and the working-tree runs, one sample at a time, for at
least ten samples. Compare the results with benchstat. Use the compare command
that AGENTS.md names. If AGENTS.md names only a plain benchmark command,
check out the base in a separate worktree and alternate by hand.
Do not compare two runs made one after the other. Drift of the machine during the first run makes a false difference in time.
If no benchmark covers the path that changed, add one in the same change. A performance change is not done until the numbers show the gain. A correctness change on a hot path reports its cost.
Report the result in the commit message
The commit message states the result in a separate paragraph, for allocations and bytes per op only, with no time figures. Allocation counts do not drift between runs, and times age with every CPU and Go release.
- No change:
Benchmarks: no change (BenchmarkFoo, BenchmarkBar). - A change, where
benchstatreports a significant difference and the values differ at three significant figures: before → after and the percentage, for each benchmark that changed. For example:BenchmarkFoo: allocs/op 8.37M → 7.10M (−15.2%), B/op 1.90 GB → 1.62 GB (−14.7%).
A change that is not measured has no benchmark paragraph.
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.
- 5d ago First seen · 48 lines · 57 tokens per session scan A 0d36c79e0471
go-benchmarks is a skill published in the GitHub repository maguro/pbf (57 stars, last pushed 2d ago), licensed Apache-2.0. It adds 57 tokens to every session and 515 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-11-01.
go-benchmarks compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| go-benchmarksmaguro/pbf | Skill | 57 | 2d ago | Apache-2.0 | A | 57 | Claude Code |
| golang-testingaffaan-m/ECC | Skill | 298k | 4d ago | MIT | A | 37 | Claude Code, Kiro |
| golang-patternsaffaan-m/ECC | Skill | 298k | 4d ago | MIT | A | 45 | Claude Code, Kiro |
| golang-testingsamber/cc-skills-golang | Skill | 3.8k | 5d ago | MIT | A | 115 | Claude Code |
| golang-uber-fxsamber/cc-skills-golang | Skill | 3.8k | 5d ago | MIT | A | 122 | Claude Code |
| golang-samber-oopssamber/cc-skills-golang | Skill | 3.8k | 5d ago | MIT | A | 74 | Claude Code |
| ax-go-agent-optimizeax-llm/ax | Skill | 3.0k | 3d ago | Apache-2.0 | A | 51 | Claude Code, Codex |