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 mthines/agent-skills --skill polishgit clone --depth 1 https://github.com/mthines/agent-skillsWrote 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/mthines/agent-skills/polish)<a href="https://agentmods.dev/skills/mthines/agent-skills/polish"><img src="https://agentmods.dev/badge/skills/mthines/agent-skills/polish.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Anti-Refusal · line 106 Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.Fix: Remove instructions that suppress warnings, disclaimers, or ethical commentary. Let the agent surface safety-relevant caveats to the user.
- medium Prompt Injection · line 50 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 62 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
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 | $0.00217 | $0.03801 |
| Opus 5 | $0.00109 | $0.01901 |
| Sonnet 5 | $0.00043 | $0.00760 |
| Haiku 4.5 | $0.00022 | $0.00380 |
Grade A, and why
polish 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 7d 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 — 269 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Polish — Re-runnable Branch Quality Gate
Get a branch into clean, reviewable shape before it goes up for review — and run it again any time you've made a lot of changes and want to tidy up.
This skill is an orchestrator. It does not contain quality rules of its own; it composes two existing pieces over the current branch diff:
- The
pr-revieweragent — broad review via the branch's open PR (read-only; findings are surfaced for you to act on, applied viaimplement-suggestion). Requires an open PR —pr-reviewerhas no PR-less mode. If no PR is open, Pass A is skipped and you are told to open a draft PR via/create-prfirst. - The
code-qualityskill insimplifymode — applies Class M mechanical refactors one at a time behindSkill("confidence", "code") ≥ 90 %and a scoped fast-check, reverting any that fail. Class J (judgment) recipes stay as proposals.
/create-pr delegates its post-draft quality step to review-loop (the bounded convergence loop), which calls only polish simplify — never full polish. You can run polish standalone at any point for a pre-draft local check.
Modes
Parse the first token of $ARGUMENTS. Everything else is a flag.
| Mode | Trigger | What runs |
|---|---|---|
| full (default) | No mode token | review pass (pr-reviewer, read-only), then simplify pass. The "do the works" button. |
review |
First token review |
pr-reviewer pass only — read-only review surfaced as findings for you to act on. |
simplify |
First token simplify |
code-quality simplify pass only — apply Class M mechanical refactors. |
optimize |
First token optimize |
Optimality pass only — Skill("optimize-approach", "apply") over the branch diff (gated approach rewrite). |
quick |
First token quick |
Light mechanical pass only (comments, naming, dead code). No pr-reviewer pass, no structural refactors. |
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.
- 7d ago First seen · 269 lines · 217 tokens per session scan A 11a6bace5db6
polish is a skill published in the GitHub repository mthines/agent-skills (12 stars, last pushed 2d ago), licensed MIT. It adds 217 tokens to every session and 3,801 once invoked, about $0.0011 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
pre-pr-audit
Pre-PR confidence audit with 5-dimension scoring. Use when: final check before commit/push/PR, evaluating PR readiness, assessing test quality + risk + coverage holistically. Triggers: pre-pr, readiness check, confidence audit, final verification, ready to PR, how confident. Not for: code review (use…
code-review
Review .NET changes for bugs, regressions, architectural drift, missing tests, incorrect async or disposal behavior, and platform-specific pitfalls before you approve or merge them. USE FOR: reviewing a pull request or patch in a .NET repository; checking for behavioral regressions, API misuse, or missing tests…
mine-plan-review
Strictly review one implemented repository plan against AGENTS.md, architecture, the immutable plan, implementation commits, real runtime behavior, verification evidence, and downstream contracts. Use when the user invokes the host-specific mine-plan-review skill with a plan path, asks whether a plan can be accepted…
review
Review recent changes - run all validators and report status.
verification-loop
Unified verification engine for Python data science projects. Covers environment checks, type checking, linting, tests, security scans, code review with DS anti-patterns, and notebook checks. Commands (/verify, /quality-gate) invoke different subsets of this skill.
verify-pr
Heavyweight PR verification that produces a structured report: code review (delegates to miguel-review), test verification with gap analysis and execution, and upstream assumption validation. Use when the user mentions verify PR, verify my PR, PR verification, test plan for PR, or validate PR changes.