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 agentmods add commands/jjmartres/ai-coding-agents/speckit.polishgit clone --depth 1 https://github.com/jjmartres/ai-coding-agentsWrote 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/commands/jjmartres/ai-coding-agents/speckit.polish)<a href="https://agentmods.dev/commands/jjmartres/ai-coding-agents/speckit.polish"><img src="https://agentmods.dev/badge/commands/jjmartres/ai-coding-agents/speckit.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 | $0.00018 | $0.00563 |
| Opus 5 | $0.00009 | $0.00282 |
| Sonnet 5 | $0.00004 | $0.00113 |
| Haiku 4.5 | $0.00002 | $0.00056 |
Grade A, and why
speckit.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 6d 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 — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Strictly follow the project constitution.
Apply the following skills where relevant:
- document-code
- document-project
- glab
- humanizer
- karpathy-guidelines
- mermaid-diagrams
- writing-clearly-and-concisely
Steps
1. Load project context
- Read
.specify/memory/constitution.mdif it exists - Let the constitution override any defaults below
2. @debugger run the test suite
- Execute
make test-in-ciwith coverage report - Surface all failures with file, line, and error message
- For each failure: diagnose root cause and propose a concrete fix
- Do NOT proceed to step 3 until the human confirms or fixes are applied
- Report coverage percentage and any skipped tests
3. @code-reviewer audit code documentation
- Identify undocumented or poorly documented public interfaces, functions, and modules
- Apply
document-codeskill to fill gaps - Flag anything that requires human input (ambiguous intent, missing domain context)
4. @documentation-engineer regenerate project documentation
- Apply
document-projectskill to update or generate the full documentation structure - Refresh architecture diagrams with
mermaid-diagramsskill where relevant - Ensure prose is clear and concise per
writing-clearly-and-conciselyskill
5. @documentation-engineer update CHANGELOG.md
- Follow Keep a Changelog format
- Add an entry under
[Unreleased]summarising changes since the last release - Group by: Added / Changed / Fixed / Removed
6. @ai-engineer update AGENTS.md
- Reflect any new agents, skills, tools, or behavioural constraints introduced in this iteration
7. Update Graphify knowledge graph (if applicable)
- Check whether a
graphify-out/directory exists at the project root - If it exists, run:
graphify update . --out ./graphify-out - If not, skip silently
8. Sync memory bank
- Run
/memory-bankto persist updated project state
9. Commit
- Run
/commitwith a conventional commit message scoped to this workflow
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.
- 6d ago First seen · 76 lines · 18 tokens per session scan A fcea4eda3288
speckit.polish is a command published in the GitHub repository jjmartres/ai-coding-agents (45 stars, last pushed 2mo ago), licensed MIT. It adds 18 tokens to every session and 563 once invoked, about $0.0001 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 2026-08-30.
Other commands, from other repositories
compete
NEVER workspace/[name].md (wrong) ALWAYS workspace/company/[name].md (correct).
ponytail
Switch ponytail intensity level (lite/full/ultra/off).
research-topic
Research a topic using multi-agent-researcher skill.
e2e
Generate and run end-to-end tests with Playwright. Creates test journeys, runs tests, captures screenshots/videos/traces, and uploads artifacts.
close-issue
../../knowledge/procedures/close-issue-procedure.md.
python-review
Comprehensive Python code review for PEP 8 compliance, type hints, security, and Pythonic idioms. Invokes the python-reviewer agent.