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 yuri-semenenko/ai-engineering-workspace --skill pr-recheckgit clone --depth 1 https://github.com/yuri-semenenko/ai-engineering-workspaceWrote 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/yuri-semenenko/ai-engineering-workspace/pr-recheck)<a href="https://agentmods.dev/skills/yuri-semenenko/ai-engineering-workspace/pr-recheck"><img src="https://agentmods.dev/badge/skills/yuri-semenenko/ai-engineering-workspace/pr-recheck/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/yuri-semenenko/ai-engineering-workspace/pr-recheck"><img src="https://agentmods.dev/badge/skills/yuri-semenenko/ai-engineering-workspace/pr-recheck.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 | $0.00076 | $0.02171 |
| Opus 5.5 | $0.00030 | $0.00868 |
| Sonnet 5.5 | $0.00015 | $0.00434 |
| Haiku 4.5 | $0.00008 | $0.00217 |
Grade A, and why
pr-recheck scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PR recheck
Follow-up pass on a PR that has already been reviewed once. Assumes a prior round of inline comments exists and the author has pushed fixes since.
Preconditions
The same checks as pr-review: run gh auth status; require a PR number/URL or a
checked-out branch with an open PR; verify open state via
gh pr view <id> --json state,mergedAt. One difference: if it is merged or closed,
surface that and stop rather than asking, because there is no thread left to
resolve and nothing to approve.
Steps
- Load prior threads. Fetch open review threads via GraphQL — REST does not
expose thread resolution state:
Pagegh api graphql -f query='query($owner:String!,$repo:String!,$num:Int!,$cursor:String){ repository(owner:$owner,name:$repo){ pullRequest(number:$num){ reviewThreads(first:100,after:$cursor){ pageInfo{hasNextPage endCursor} nodes{ id isResolved comments(first:20){ pageInfo{hasNextPage endCursor} nodes{ path line body author{login} } } } } } } }' \ -f owner=OWNER -f repo=REPO -F num=NreviewThreadsuntilhasNextPageis false, passing eachendCursorascursor. For every open thread, page itscommentsconnection vianode(id: $threadId) { ... on PullRequestReviewThread { comments(first: 100, after: $cursor) { pageInfo { hasNextPage endCursor } nodes { body author { login } } } } }until complete. Retain the first comment author and the authenticated user's replies with bodies, including replies after the first page. If any page is unavailable or incomplete, report that evidence gap; do not infer that the user did not accept a risk or mark it accepted without evidence. Keep only threads withisResolved: false. - Load new work. Diff and commits since the previous review (
gh pr diff <id>plus the commit log). Read the new code fully, cross-referencing touched files. - Build the verification matrix. For each open thread, record three separate
things, with the file:line and the commit that addresses it. Output as a table.
- Severity, from the first review: Critical / Important / Optional. When the thread does not state it, classify it from what it describes under pr-review rules and mark the severity as inferred. When it reads as either of two tiers, take the higher one: severity alone decides what blocks, and pr-review posts comments as prose that rarely names a tier.
- Status:
addressed/partial/not addressed/wont-fix (author replied). Awont-fixreply is the author's position, not the reviewer's acceptance.addressed (not resolved)marks anaddressedthread another reviewer opened, which the decision gate leaves open for them; it does not block. - Blocks approval:
yes,no, orno (accepted risk). A Critical or Important thread blocks unless it isaddressedor its risk is accepted. An Optional thread does not block.
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.
- 2d ago Changed 2457f9e45cb8
- 4d ago Changed · +26 lines 504235ba0829
- 14d ago Changed · +38 lines · -33 tokens per session cdb4c0fbdda3
- 2mo ago First seen · 79 lines · 109 tokens per session scan A 33ef270c7923
pr-recheck is a skill published in the GitHub repository yuri-semenenko/ai-engineering-workspace (4 stars, last pushed 4d ago), licensed MIT. It adds 76 tokens to every session and 2,171 once invoked, about $0.0003 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 2026-08-31.
pr-recheck compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| pr-recheckyuri-semenenko/ai-engineering-workspace | Skill | 4 | 4d ago | MIT | A | 76 | Claude Code |
| 056-design-avoid-breaking-changesjabrena/plinth | Skill | 462 | 3d ago | Apache-2.0 | A | 106 | Claude Code, Codex |
| shiprjmurillo/ai-agents | Skill | 49 | today | MIT | A | 80 | Claude Code |
| git-advanced-workflowsrjmurillo/ai-agents | Skill | 49 | today | MIT | A | 90 | Claude Code |
| ai-prarcasilesgroup/ai-engineering | Skill | 65 | 5d ago | Apache-2.0 | A | 60 | Claude Code, Codex |
| maintainer-reviewrjmurillo/ai-agents | Skill | 49 | today | MIT | A | 73 | Claude Code |
| chestertons-fencerjmurillo/ai-agents | Skill | 49 | today | MIT | A | 80 | Claude Code |