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.
git clone --depth 1 https://github.com/ThibautBaissac/rails_ai_agentsWrote 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/thibautbaissac/rails_ai_agents/pr-artifact)<a href="https://agentmods.dev/commands/thibautbaissac/rails_ai_agents/pr-artifact"><img src="https://agentmods.dev/badge/commands/thibautbaissac/rails_ai_agents/pr-artifact/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/thibautbaissac/rails_ai_agents/pr-artifact"><img src="https://agentmods.dev/badge/commands/thibautbaissac/rails_ai_agents/pr-artifact.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 | $0.00032 | $0.02944 |
| Opus 5 | $0.00016 | $0.01472 |
| Sonnet 5 | $0.00006 | $0.00589 |
| Haiku 4.5 | $0.00003 | $0.00294 |
Grade A, and why
pr-artifact 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 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.
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 — 331 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Presenting Pull Requests
Create one unique, self-contained HTML artifact that helps the intended audience understand what changed, why it changed, how the change works, what is affected, how it was validated, and what risks or questions remain.
Input
PR or change source: $ARGUMENTS
Accept any of these input forms:
- PR URL.
- PR number.
- Branch name, commit range, or comparison ref.
- Local diff or patch file.
- Markdown summary, release note, review notes, or implementation notes.
- Empty input, if the current repository and branch can provide enough context.
If $ARGUMENTS is empty, first try to infer the change from the current git repository:
- Identify the current branch.
- Compare it against the repository's default branch when that is locally available.
- If no reliable comparison base exists, ask the user for a PR URL, PR number, branch, commit range, or diff file and stop.
If $ARGUMENTS references network-only PR metadata and network access is unavailable, use local git information when possible and clearly mark unavailable metadata under Open questions.
Output Requirement
Create a single .html file inside the artifacts/ directory in the current working directory.
If artifacts/ does not exist, create it.
The file must be fully self-contained:
- Include all CSS in a
<style>tag. - Include all JavaScript in a
<script>tag. - Do not depend on external CDNs, fonts, images, packages, analytics, or network requests.
- Use semantic HTML and accessible markup.
- Work by opening the file directly in a browser.
Name the file using a short slug derived from the PR title, branch, or change theme, for example:
artifacts/pr-artifact-[slug].html
If no clear title exists, use:
artifacts/pr-artifact.html
Do not overwrite an existing unrelated artifact unless the user explicitly asks. If the target filename exists, choose a clear variant such as artifacts/pr-artifact-[slug]-v2.html.
Core Principles
- Treat the PR, diff, commits, and source notes as authoritative.
- Explain the change at the right altitude for reviewers and stakeholders.
- Preserve technical accuracy while translating dense implementation details into plain language.
- Make the review surface obvious: affected areas, behavior changes, validation, risk, rollout, and open questions.
- Separate verified facts from inferred implications.
- Make missing details visible instead of filling gaps with confident guesses.
- The artifact should feel intentionally designed, not templated.
- Avoid generic dashboard filler, decorative-only graphics, excessive cards, repeated section layouts, and one-note color palettes.
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 · 331 lines · 32 tokens per session scan A fac5454ebb49
pr-artifact is a command published in the GitHub repository ThibautBaissac/rails_ai_agents (661 stars, last pushed 3mo ago), licensed MIT. It adds 32 tokens to every session and 2,944 once invoked, about $0.0002 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
report
Render a consolidated customer-review Excel workbook, optional audit-grade PDF, and chain-of-custody manifest from approved findings. Gated on ≥1 approve in feedback.jsonl.
done
Finish a task - document, create PR or merge, close.
latex-diff
Compare two LaTeX files, project directories, or Git revisions and produce human-readable plus machine-readable severity-graded semantic changes. Use when determining what changed between manuscript versions or preparing a revision audit bundle. Not for proofreading one version; use $proofread.
review-export
Export pipeline review with resume-export-expert (PDF/DOCX rendering, contract, golden parity).
review-template
Resume template review with resume-export-expert (rendering impl by pdf-docx-generator).
audit
Audit the premises of a document — refute-first, one claim per agent, refutations returned as checkable anchors.