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 skills add D-Robotics/moss --skill bench-reprogit clone --depth 1 https://github.com/D-Robotics/mossWrote 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/d-robotics/moss/bench-repro)<a href="https://agentmods.dev/skills/d-robotics/moss/bench-repro"><img src="https://agentmods.dev/badge/skills/d-robotics/moss/bench-repro/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/d-robotics/moss/bench-repro"><img src="https://agentmods.dev/badge/skills/d-robotics/moss/bench-repro.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 | $0.00020 | $0.00305 |
| Opus 5.5 | $0.00008 | $0.00122 |
| Sonnet 5.5 | $0.00004 | $0.00061 |
| Haiku 4.5 | $0.00002 | $0.00030 |
Grade A, and why
bench-repro 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 23d 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
Reproducing bench evidence
- Internal bench:
npm run bench -- --task <id> --samples <n> --label <label>(provider viaMOSS_BENCH_API_KEY,--model,--base-url; temp=0; approval=never). Results land inbench/results/<label>/(never committed). - A/B:
npm run bench:ab -- <engine> --samples <n>forbest-of-n/reasoning-high/goal-loop/model-routing; the verdict json lands inbench/results/ab-<engine>-<stamp>.json. - Noise discipline: compare only same-SHA runs; use
npm run bench:noisefor the noise band; a delta inside the band is not evidence. - SWE-bench (external board):
npm run bench:swe -- --samples 2 --label <label>produces predictions; grade withnpm run bench:swe -- --eval --label <label>. Instance subset is locked inbench/boards/swebench-instances.json— never edit it mid-study. - A baseline must be collected from a pinned build (
--dist bench/.cache/<snapshot>/dist), not from a dirty tree.
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.
- 23d ago First seen · 14 lines · 20 tokens per session scan A 5e196f1ca11d
bench-repro is a skill published in the GitHub repository D-Robotics/moss (143 stars, last pushed 5d ago), licensed MIT. It adds 20 tokens to every session and 305 once invoked, about $0.0001 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 2026-10-01.
Other skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.
skill-benchmark
Benchmark AI skill effectiveness by measuring implementation quality against legacy constraints.