Borrowing it
Nothing to install: this file belongs to alfredoperez/speckit-companion. 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/alfredoperez/speckit-companion/main/.claude/commands/bench-capture.mdgit clone --depth 1 https://github.com/alfredoperez/speckit-companionWrote 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/alfredoperez/speckit-companion/bench-capture)<a href="https://agentmods.dev/commands/alfredoperez/speckit-companion/bench-capture"><img src="https://agentmods.dev/badge/commands/alfredoperez/speckit-companion/bench-capture/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/commands/alfredoperez/speckit-companion/bench-capture"><img src="https://agentmods.dev/badge/commands/alfredoperez/speckit-companion/bench-capture.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 | $0.00015 | $0.00866 |
| Opus 5 | $0.00008 | $0.00433 |
| Sonnet 5 | $0.00003 | $0.00173 |
| Haiku 4.5 | $0.00002 | $0.00087 |
Grade A, and why
bench-capture 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 — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Your task
After a size has been run through its three cells, capture the evals, append to history, regenerate the report, and reset the cells for the next round.
1. Resolve the size
From $ARGUMENTS — easy | medium | hard | oversized. If missing, ask with AskUserQuestion.
2. Read the cell mapping
node ../speckit-bench/run-all.mjs --dry-run
It prints which letter carries which arm, plus the spec-kit, extension and Companion versions this round was measured with. You need the mapping to attribute the reviews; the judges below must not be told which arm they are looking at.
3. Independent rubric judges (before measuring — they read the diff)
Spawn one judge agent per cell (a parallel() Workflow of 3). Each judge — which did NOT write the code — does exactly:
- Read the diff:
git -C ~/dev/projects/conduit-<size>-<letter> diff bench-baseline. - Read the requirements at
../speckit-bench/prompts/conduit/<size>.mdand the conventions in the cell'sCLAUDE.md. - Score 1–5:
readability,conventions,scope. Judge the layering rules the app actually documents; do not penalize a rule the size had no occasion to exercise. - Write
quality = {readability, conventions, scope, justification}into../speckit-bench/runs-meta/conduit-<size>-<letter>.json(MERGE — keeprunId/size/mode).
Give each judge the letter only. A judge told it is looking at "the Companion arm" is no longer independent.
4. Comparative reviewer (one agent, cross-solution)
The per-cell judges score each solution in isolation, and tests cannot fully pin correctness. Spawn one reviewer that sees all three at once. Run it before step 5 — measuring resets the cells and erases the diffs. It does exactly:
- Read all three diffs (
git -C ~/dev/projects/conduit-<size>-<letter> diff bench-baseline), labeled A, B and C. - Read the requirements and the conventions.
- Produce a comparative review:
- Ranking best to worst, one line each.
- Head-to-head differences — structure, layering, naming, edge cases, test coverage.
- Suspected bugs the deterministic harness cannot catch, per solution (or "none found").
- One-line verdict per solution.
- Write it to
../speckit-bench/reviews/<size>.md— committed there, accumulating: read the existing file first, then prepend a## <YYYY-MM-DD> — <size>section so prior runs are preserved.
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 Changed · +3 lines · -20 tokens per session 2b4b23c7d876
- 12d ago First seen · 54 lines · 35 tokens per session scan A 8724908ce39b
bench-capture is a command published in the GitHub repository alfredoperez/speckit-companion (90 stars, last pushed today), licensed MIT. It adds 15 tokens to every session and 866 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 2026-08-30.
Other commands, from other repositories
pr-ready
Run the project's pre-commit review loop to determine whether the current branch is ready to push — lint, tests, parallel pr-review-toolkit agents plus an over-engineering audit, fix-and-re-run until convergence.
devs-team
Parallel multi-lens critique of a solution design (the active spec if present, else the decision brief). Dispatches 5 engineering lenses, merges findings, reports a verdict. Report-only, never blocks.
quiz-gate
The ★-tap NUDGE before merging a significant+worthy gate PR: a 5-question quiz built from the ACTUAL diff+tests, handed to the operator's configured teacher seam. Three logged responses (engage/defer/wave); advisory, never must-pass. Gates the human's attention, not the merge.
review
Multi-agent code review with parallel validation.
pr-feedback-ingest
Turn PR feedback (Greptile, CI, bots, humans) into a structured traceable backlog aligned with TASKSTATE.md, DECISIONS.md, and IMPLEMENTATIONPLAN.md so the next execution step can be a narrow implement-approved-slice or a small planning touch without losing alignment. Corrective scope only. Use when a PR is open or…
review-hard
Review the current task changes for real correctness, safety, and maintainability risks before slice closure or PR prep, and recommend the smallest safe next step. Surfaces meaningful issues (not cosmetic feedback); not a replacement for external review systems. Returns no-op when the review would not materially…