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 IgorWarzocha/howaboua-pi-stuff --skill skill-creatorgit clone --depth 1 https://github.com/IgorWarzocha/howaboua-pi-stuffWrote 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/igorwarzocha/howaboua-pi-stuff/skill-creator)<a href="https://agentmods.dev/skills/igorwarzocha/howaboua-pi-stuff/skill-creator"><img src="https://agentmods.dev/badge/skills/igorwarzocha/howaboua-pi-stuff/skill-creator/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/igorwarzocha/howaboua-pi-stuff/skill-creator"><img src="https://agentmods.dev/badge/skills/igorwarzocha/howaboua-pi-stuff/skill-creator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00015 | $0.00782 |
| Opus 5.5 | $0.00006 | $0.00313 |
| Sonnet 5.5 | $0.00003 | $0.00156 |
| Haiku 4.5 | $0.00002 | $0.00078 |
Grade A, and why
skill-creator 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 2mo 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.
What it actually says
- For revisions, read the existing
SKILL.mdand every file it directly loads before changing structure or removing material. - Check the available skills before drafting. Read any whose guidance materially applies to the domain, artifact, or writing mode of the skill being authored.
- Choose the package scope before drafting. A global skill must apply across workspaces in the user's environment. It may be specialist, but must not assume one repository's paths, packages, commands, versions, provider payloads, or internal lifecycle. Distill project evidence into portable decisions, invariants, and failure modes. Keep project mechanics in a project-scoped skill or guidance. A project skill may be a delta-only addendum: give it a distinct name, tell the agent to discover and load an applicable general skill, then state only local constraints, exceptions, paths, and completion. A project skill with the same name as a global skill shadows it, so use that name only for an intentional complete replacement. Check the global catalog before naming. Do not copy or hard-code the global package.
- Ground the work in two or three real requests, observed failures, or explicit user corrections. Identify the decision, failure, or quality bar the skill must change.
- Write the description as the shortest reliable call trigger, not a summary of the package.
- Address the reading agent as you at selection time. Say when to read the skill or what behaviour now applies.
- Prefer direct forms such as
Read before...,Use for..., orMust always apply. - Do not describe what the skill offers, improves, helps with, or contains.
- Add only the boundary needed to avoid the nearest real collision.
- Keep package contents, routes, examples, and rationale in the body.
- Do not pad with prompt synonyms or capability inventory.
- Quote it. Aim for 175 characters or fewer. Characters 176 through 200 require justification. More than 200 is an error.
- Start every skill and reference body inside execution. Do not add a title, filename heading, purpose, activation, or other preamble that repeats its trigger or route.
- Keep the normal path and judgment needed on every run in
SKILL.md. Put branch-specific detail behind exact reference paths. Use scripts only where deterministic code is simpler and more reliable than prose. - When one skill needs another capability, tell its reader to discover and load an available applicable skill for that capability. Never hard-code another package's name because skill inventories differ between environments. Name it as a skill, for example
load an applicable review skill, notload code-revieworuse "code review". - Keep a line only when it changes a decision, prevents a known failure, preserves user-authored intent, or defines observable completion. Remove model defaults, host guarantees, generic encouragement, duplicate checks, and facts cheaply recovered from the environment.
- Match the body's register to how language affects execution.
- Operational, coding, tooling, review, and SOP skills use terse instructions, explicit conditions, exact checks, and observable boundaries.
- Creative, voice, taste, and ideation skills may use evocative, opinionated, or playful language when that language directly steers the output. Their call triggers remain direct and terse.
- For revisions, preserve verified domain judgment and useful failure handling. Replace stale or conflicting guidance instead of appending history.
- Exercise bundled scripts and verify every referenced path. Review call and near-collision requests only when the trigger changed materially or an observed selection failure warrants it. Do not build synthetic suites that merely prove the skill exists.
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.
- 2mo ago First seen · 28 lines · 15 tokens per session scan A ed06474a5861
skill-creator is a skill published in the GitHub repository IgorWarzocha/howaboua-pi-stuff (413 stars, last pushed yesterday), licensed MIT. It adds 15 tokens to every session and 782 once invoked, about $0.0001 per session on Opus 5.5. 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.
skill-creator compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| skill-creatorIgorWarzocha/howaboua-pi-stuff | Skill | 413 | yesterday | MIT | A | 15 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 103 | Claude Code |