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 hashiiiii/rules-for-ai --skill hashiiiii-write-engit clone --depth 1 https://github.com/hashiiiii/rules-for-aiWrote 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/hashiiiii/rules-for-ai/hashiiiii-write-en)<a href="https://agentmods.dev/skills/hashiiiii/rules-for-ai/hashiiiii-write-en"><img src="https://agentmods.dev/badge/skills/hashiiiii/rules-for-ai/hashiiiii-write-en/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/hashiiiii/rules-for-ai/hashiiiii-write-en"><img src="https://agentmods.dev/badge/skills/hashiiiii/rules-for-ai/hashiiiii-write-en.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 | $0.00026 | $0.00574 |
| Opus 5 | $0.00013 | $0.00287 |
| Sonnet 5 | $0.00005 | $0.00115 |
| Haiku 4.5 | $0.00003 | $0.00057 |
Grade A, and why
hashiiiii-write-en 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 — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
English Writing
Write a simple draft. Then revise its language and structure. Keep the requested audience, tone, format, and length.
Meaning
- Preserve facts, quantities, conditions, exceptions, uncertainty, and requirement strength. Do not turn a limited observation into a general claim.
- Do not invent causes, sources, opinions, or personal experience.
- Keep code, commands, identifiers, paths, URLs, quotations, and product names exact unless the task requires their change.
- Use one term per meaning. Do not merge distinct concepts to reduce vocabulary.
Draft
- Start with the useful point. Give each sentence one main idea and each paragraph one topic.
- Use familiar words and complete grammar. If the reader needs help with a technical term, explain it.
- Use verbs for actions. Break long noun chains into phrases with prepositions.
- Do not invent compounds with hyphens to shorten explanations. Use verbs or prepositions, such as "data serialized as text" instead of "text-serialized data". Keep established terms and hyphens needed to prevent misreading.
- If the actor is known, use active voice. Systems can act. Do not invent a person to supply a subject.
- For instructions, give one action per sentence.
- In technical text, limit instructions to 20 words per sentence and explanations to 25. Count each code span as one word.
Revise
- Delete empty emphasis and stock phrases such as "seamlessly", "game-changing", and "it is worth noting".
- Replace vague praise with supplied facts. If there is no fact to retain, delete the praise.
- State the point directly. Remove decorative contrasts, rhetorical questions, repeated conclusions, and em dashes from prose.
- Keep comparisons, qualifications, and courtesy that serve the reader.
- If headings or lists help navigation or comparison, use them. Do not force three items, equal paragraphs, or punchy endings.
Check
Before delivery:
- Compare each claim with the source. Check who acts, what they act on, quantities, timing, exceptions, uncertainty, requirement strength, and exact technical text.
- Compare scope words such as "one", "some", "all", and "only". Correct any change in their meaning. Do not infer links between nearby facts.
- Check the longest sentences, terms, repeated patterns, and requested length. Correct each defect.
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 · +1 lines 0a3be60d65d8
- 3d ago First seen · 46 lines · 26 tokens per session scan A f3de37b0eb2b
hashiiiii-write-en is a skill published in the GitHub repository hashiiiii/rules-for-ai (10 stars, last pushed 3d ago), licensed MIT. It adds 26 tokens to every session and 574 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 2026-09-06.
Other skills, from other repositories
aiops
Flow Conductor for the aiops bundle. One entry — infers task type, shows plain-language steps, tracks journey in flow.state.yaml, dispatches skill + agent per phase.
review
Review in three modes — code (diff vs standards/spec), design (NOTES + tech-spec before planning), drift (implementation vs tech-spec before ship). Pick mode from Flow Conductor phase or user intent; do not mix artifact types.
to-issues
Break a plan or PRD into vertical-slice issues on the project issue tracker.
ui-mockup
Generate HTML/CSS mockups for UI design. Produces browser-previewable mockup files with placeholder data. Use when the user says ui-mockup, mockup, design UI, wireframe, or needs visual reference for a feature.
to-prd
Synthesize the current conversation into a PRD and publish to the issue tracker. No interview — use what you already know.
gitops
Git operations wrapper: sync, stage, commit, push. Generates Conventional Commits messages with .scratch artifact references. Use when the user says gitops, commit, push, sync code, or needs version control operations.