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 decocms/studio --skill jira-reviewgit clone --depth 1 https://github.com/decocms/studioWrote 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/decocms/studio/jira-review)<a href="https://agentmods.dev/skills/decocms/studio/jira-review"><img src="https://agentmods.dev/badge/skills/decocms/studio/jira-review/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/decocms/studio/jira-review"><img src="https://agentmods.dev/badge/skills/decocms/studio/jira-review.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 | $0.00041 | $0.01108 |
| Opus 5.5 | $0.00016 | $0.00443 |
| Sonnet 5.5 | $0.00008 | $0.00222 |
| Haiku 4.5 | $0.00004 | $0.00111 |
Grade A, and why
jira-review 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 13d 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 — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
jira-review — review what the other run built
Insert this into the prompt of the Jira column your team reviews in. It is the
second of two runs: an earlier column implemented the issue and opened a pull
request (jira-execute), and this one judges it.
It is ordinary text once inserted — read it, cut what does not apply to your team, add what does.
You are reviewing someone else's change to a Jira issue. The issue itself is above, web links included.
Find what you are reviewing
- The pull request is normally on the issue as a web link, put there by the run
that implemented it. If it is not, its branch is normally named for the issue
key:
gh pr list --search "<the issue key>" --state all. - If there is genuinely no pull request, do not write one. Report that on the issue and stop.
Review the code
- Check the branch out and read the diff (
gh pr diff <n>, then the files around it). Does it do what the issue asks; is it reachable from the surface the issue names; does it break something next to it. - Do NOT implement the fix yourself and do NOT open a pull request. A reviewer that rewrites the change has reviewed nothing, and a second pull request on one issue is what a human then has to clean up. A small, unambiguous correction is a comment on the issue, not a commit.
QA it on the deploy preview
- The deploy bot comments the preview URL on the pull request
(
gh pr view <n> --json comments); the issue's web links may carry it too. If no preview exists yet, poll until it does. A first request may wake a hibernating build; retry until you get real HTML. - Exercise the behaviour and MEASURE it. Read the value back from the page rather than inferring it from a name or a spec.
- Compare against production, not only against the spec: the live site is the "before" and the preview is the "after". A difference you cannot see on both is not evidence. If production is unreachable from this pod, say so in your report rather than passing the card on an untested comparison.
- Capture before/after evidence with
qa-screenshot <url> <path>.png [--mobile], on desktop AND mobile, andReadeach file. A screenshot you never opened is not verification, and mobile is not desktop resized — plenty of these issues only reproduce on one of the two.
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.
- 13d ago Changed 754c4b4c5eb7
- 15d ago Changed · +3 lines 710c7495d29d
- 29d ago Changed · +5 lines 1b7a45568a90
- 1mo ago First seen · 80 lines · 41 tokens per session scan A 64ec69b8fc4c
jira-review is a skill published in the GitHub repository decocms/studio (409 stars, last pushed today), licensed MIT. It adds 41 tokens to every session and 1,108 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 2026-09-16.
Other skills, from other repositories
otto-pr-review-loop
Run Polly and OCR on a PR until all findings settle.
resolve-drive-pr
Drive an open PR through current CI, independent Polly and OCR reviews, preview, and maintainer handoff.
resolve-review-pr
Review an existing fix PR using the audited repro and full diff, preserving branch and approval rules.
resolve-impact-assessment
Check the full candidate diff, affected consumers, focused tests, evidence identities, and verification gaps.
cross-review
Verify an implementer's diff with an INDEPENDENT, different-vendor sub-agent (diff plus contract only); turn blocking issues into fix-tasks and loop until clean.
otto-review-remediation
Handle trusted reviewers' change requests on Otto PRs.