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 rtorcato/repo-tooling --skill dogfoodgit clone --depth 1 https://github.com/rtorcato/repo-toolingWrote 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/rtorcato/repo-tooling/dogfood)<a href="https://agentmods.dev/skills/rtorcato/repo-tooling/dogfood"><img src="https://agentmods.dev/badge/skills/rtorcato/repo-tooling/dogfood/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/rtorcato/repo-tooling/dogfood"><img src="https://agentmods.dev/badge/skills/rtorcato/repo-tooling/dogfood.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 | $0.00104 | $0.02254 |
| Opus 5 | $0.00052 | $0.01127 |
| Sonnet 5 | $0.00021 | $0.00451 |
| Haiku 4.5 | $0.00010 | $0.00225 |
Grade C, and why
dogfood scanned grade C with 1 finding 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 10d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
! rm -rf <BASE> How it starts
The opening of the file, as written. The whole thing — 210 lines — stays where its author put it; the contents beside it link to each section on GitHub.
dogfood
Point the repo's own tooling at repos it has never seen and find out what it does wrong. Arguments: $ARGUMENTS
The bugs live where the tool meets content it did not write. An empty
directory finds nothing — every finding from the run this skill is based on came
from a fixture that already had a package.json, a manifest, or a source file
with an opinion in it. Scaffolding onto nothing is the one case the authors
already tested.
What this never does
- Never writes outside its temp directory. Not the repo's working tree, not
~/.claude, not a global git config (git config --globalwrites a stowed dotfile on this machine). - Never deletes.
rmis often permission-blocked for an agent, and a half-deleted fixture is worse than a kept one. Report the path and size at the end; the user removes it when they are done reading it. - Never files an issue without asking, and never labels one
ai-ready— that label is the human's gate intoai-issue-loop.
Step 1 — ask what to exercise
Use AskUserQuestion. Look at the repo first so the options are real — read its
package.json bin, its CLI's --help, or its presets/templates directory —
then ask:
- What to exercise. Offer the actual entry points found (multiSelect). For a scaffolding tool that is its presets; for a linter its rule sets; for a codemod its transforms.
- Fixture shape. Realistic pre-existing repos (recommended — this is what finds bugs) vs empty directories (only worth it to check the happy path still works).
- What to do with findings. Report in the transcript only (recommended
for a first run) vs also file GitHub issues. If they choose issues, every
one opens with
🤖 *Filed by an agent via dogfood.*and carries noai-readylabel.
Skip a question the arguments already answer.
Step 2 — pin the build and the version
A finding with no version stamp is unreproducible and will be argued with.
Build from source, and record the commit — not the version in package.json,
which under semantic-release without @semantic-release/git never moves:
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.
- 10d ago First seen · 210 lines · 104 tokens per session scan C 0ac1b26ad13d
dogfood is a skill published in the GitHub repository rtorcato/repo-tooling (2 stars, last pushed 9d ago), licensed MIT. It adds 104 tokens to every session and 2,254 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
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".
neuron-evaluation-engineer
Create and run AI evaluations with datasets, assertions, and output drivers in Neuron AI. Use this skill whenever the user mentions evaluation, testing AI systems, creating evaluators, dataset-driven testing, assertion-based validation, or wants to measure AI system performance. Also trigger for tasks involving…
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.