Borrowing it
Nothing to install: this file belongs to frostney/pascal-mcp-sdk. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/frostney/pascal-mcp-sdk/main/.agents/skills/review-pr/SKILL.mdgit clone --depth 1 https://github.com/frostney/pascal-mcp-sdkWrote 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/frostney/pascal-mcp-sdk/review-pr)<a href="https://agentmods.dev/skills/frostney/pascal-mcp-sdk/review-pr"><img src="https://agentmods.dev/badge/skills/frostney/pascal-mcp-sdk/review-pr.svg" alt="Measured on agentmods" 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 | $0.00047 | $0.00908 |
| Opus 5 | $0.00023 | $0.00454 |
| Sonnet 5 | $0.00009 | $0.00182 |
| Haiku 4.5 | $0.00005 | $0.00091 |
Grade A, and why
review-pr 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 8d 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 — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review PR
Resolve the current PR's actionable review findings without creating a second
review conversation. With the exact automatic-merge qualifier, converge the
PR and merge it.
Invariants
- Preserve unrelated work. Never amend, force-push, or revert changes you did not author.
- Reply only in the originating review thread when a reply is useful. Do not
post top-level PR summaries or issue comments. In
automatic-mergemode, a documented review-tool retrigger command is the only allowed top-level comment. - Validate findings before changing code and run the relevant project checks after fixes.
- Discover active review tools from current repository configuration, branch protection, checks, and PR activity. Do not hardcode one provider or require an integration that is disabled, historical, or merely installed.
- Treat review, approval, and CI evidence as valid only for the current PR head. A new commit or baseline merge resets every affected gate.
Automatic merge
The exact automatic-merge qualifier authorizes relevant fixes, validation,
new commits, plain pushes, documented reviewer retriggers, monitoring, one
squash merge, source-branch deletion, and local cleanup under git-workflow.
Normal /review-pr remains non-merging. An explicit read-only instruction
remains non-mutating and disables automatic merge.
An active review tool is a merge gate when repository policy or the current PR shows it was intentionally invoked. Inspect inline threads plus top-level reviews, summaries, suggestions, and nitpicks. For CodeRabbit when active, a rate limit, quota response, incomplete run, or missing verdict is pending rather than passed; apply equivalent evidence-based semantics to other tools.
Workflow
- Confirm the branch has an open PR and merge the remote default branch if behind.
- Read the current head, PR diff, unresolved threads, top-level review findings, active review tools and their terminal states, required checks, affected code, and applicable project instructions.
- If
resolve-reviewsis registered, use it for thread mechanics while keeping this skill's invariants. Otherwise handle the threads directly. - Evaluate every current finding. Fix validated in-scope findings; reply inline when acknowledgement, clarification, or a question is needed; resolve completed threads. Dismiss invalid, obsolete, duplicate, or out-of-scope findings only with evidence. Never silently ignore a nitpick.
- Run checks relevant to the changed behavior, including rendered UI and accessibility checks for user-facing changes.
- Use
/update-prto commit and push. If unavailable, follow its documented no-amend, no-force-push workflow directly. - In normal mode, report the outcome, findings addressed, commits, observed validation, review-tool state, and PR URL.
- In
automatic-mergemode, repeat the workflow against the new head. Wait with bounded backoff for required checks and every active review tool. When a tool is incomplete, errored, or rate-limited, use its documented retrigger mechanism when current evidence says it is allowed; a required command comment may use the narrow top-level exception. Do not impose a retry-count limit beyond the host platform. - Stop without merging for a material product decision, unrelated failure, unsafe or divergent PR, unavailable terminal external dependency, or unresolved required finding. Report the exact blocker.
- Once the current head is ready, every required check is green, every active
review tool has a completed verdict, and no actionable finding remains,
squash-merge and delete the source branch through
git-workflow. Sync the local default branch, remove only clean worktrees owned by this run, and report the merged PR, final head, validation, reviews, and cleanup.
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.
- 8d ago First seen · 82 lines · 47 tokens per session scan A e6741f9cadf8
review-pr is a skill published in the GitHub repository frostney/pascal-mcp-sdk (2 stars, last pushed 12d ago), licensed MIT. It adds 47 tokens to every session and 908 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 2026-08-31.
Other skills, from other repositories
api-linter
MCP definition linter rules reference. Use when bun run lint:mcp or bun run devcheck reports a lint error or warning (format-parity, schema-is-object, name-format, server-json-, etc.) and you need to understand the rule, its severity, and how to fix it. Every rule ID the linter emits has an entry in this doc.
code-simplifier
Post-session code review and cleanup against a working tree of changes. Analyzes git diff to simplify, consolidate, and align changed code with the existing codebase — modernize syntax, remove unnecessary complexity, consolidate duplicated logic, catch efficiency issues. Use after a substantive working session, or…
code-review
Reviews a supplied code path or diff for correctness, security, maintainability, and style without executing or modifying it.
remarc
Manage Remarc session comments and contextual feedback. Use when the user mentions Remarc sessions, comments, handoff, triage, review, addressing, resolving, status updates, or summaries.
code-review
Review ServiceNow server-side scripts for ES5 violations, ACL/injection/XSS issues, N+1 queries, missing setLimit/error handling, hard-coded sysids, and business-rule recursion risks.
mcp-review
Review MCP server specifications and updates for compliance, security, and quality. Use when evaluating server.json files, PRs adding/updating servers, or assessing MCP server changes. NOT for creating new entries (use add-mcp-server instead).