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 takubii/smithy --skill polishgit clone --depth 1 https://github.com/takubii/smithyWrote 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/takubii/smithy/polish)<a href="https://agentmods.dev/skills/takubii/smithy/polish"><img src="https://agentmods.dev/badge/skills/takubii/smithy/polish.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 | $0.00048 | $0.00484 |
| Opus 5 | $0.00024 | $0.00242 |
| Sonnet 5 | $0.00010 | $0.00097 |
| Haiku 4.5 | $0.00005 | $0.00048 |
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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
polish
Run an independent review protocol after implementation.
Run condition
Run after implementation and before final approval or PR unless the change is trivial, such as docs-only edits, renames, or an obvious one-file fix.
Review delegation
For non-trivial changes, use Task to run an independent reviewer. The reviewer should inspect the change against the intent and return findings only. Do not ask the reviewer to apply fixes or make adoption decisions.
Reviewer input
Give the reviewer:
- User intent.
- Acceptance criteria.
- Non-goals.
- Current diff.
- Relevant files or tests.
- Verification already run.
- Previously rejected findings and reasons, if any.
Do not give:
- The implementer's detailed rationale.
- A defense of the chosen design.
- The expected answer.
- Unnecessary conversation history.
Reviewer prompt shape
Ask for material findings only:
Review this change against the stated intent.
Intent:
...
Acceptance criteria:
...
Non-goals:
...
Verification already run:
...
Review target:
<diff and relevant files>
Focus on:
1. Behavior that does not satisfy the intent.
2. Missing edge cases or tests.
3. Unnecessary or unrelated changes.
4. Inconsistency with nearby code or existing conventions.
Do not suggest broad refactors outside the intent.
Do not re-litigate non-goals or previously rejected findings unless new evidence changes the premise.
Return findings only. If there are no material findings, say so.
For each finding include severity, evidence, why it matters, and suggested fix.
Output format
Return:
## Findings
- Severity:
Evidence:
Why it matters:
Suggested fix:
## Review recommendation
- Adopt / reject / defer suggestions are advisory only.
- Recommend another round only if material changes are made.
## Convergence
- No material findings / needs adjudication / review cap reached.
Boundaries
Do not apply fixes. Do not make final adoption decisions. Verify every finding against primary sources before acting.
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 · 98 lines · 48 tokens per session scan A 2cdf2c358080
polish is a skill published in the GitHub repository takubii/smithy (2 stars, last pushed 2mo ago), licensed MIT. It adds 48 tokens to every session and 484 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
flow-next-impl-review
Carmack-level implementation review of changes via the configured backend. Use when asked to review code or a diff in a flow-next repo.
flow-next-spec-completion-review
Verify that a spec's completed tasks fully implement the spec requirements. Use at spec completion before close.
flow-next-plan-review
Carmack-level review of a flow-next spec or plan via the configured backend. Use when asked to review a plan or spec.
flow-next-resolve-pr
Resolve PR review feedback. Fetches unresolved threads, triages, fixes, replies and resolves via GraphQL. Use when asked to address review comments.
flow-next-worktree-kit
Manage git worktrees (create/list/switch/cleanup) and copy .env files. Use for parallel feature work, isolated review, clean workspace, or when user mentions worktrees.
gate-check
Final quality gate before reporting task completion (Gate 3). Fills confidence assessment, runs supplementary verification for <95% items, completes self-check checklist. Writes PASS/FAIL result to subtask.md, then calls autoworker:dispatch for routing.