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 skills/squidbay/factory/skill-creatornpx skills add squidbay/factory --skill skill-creatorgit clone --depth 1 https://github.com/squidbay/factoryWhat 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 | $0.00097 | $0.00563 |
| Opus 5 | $0.00048 | $0.00282 |
| Sonnet 5 | $0.00019 | $0.00113 |
| Haiku 4.5 | $0.00010 | $0.00056 |
Grade A, and why
skill-creator (pointer) 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.
What it actually says
Skill Creator — mint it properly, from the source
Anthropic maintains the canonical skill-creation skill at github.com/anthropics/skills — including the format spec, authoring guidance, and evaluation tooling. This factory points at it rather than shipping a copy: the source stays current; copies rot. (Licenses in that repo are per-skill; skill-creator itself is Apache-2.0 — if your team ever vendors it into this repo, its LICENSE.txt comes along and a row goes into ../../THIRD_PARTY_NOTICES.md in the same PR.)
When a skill should exist
The third time the team does the same kind of task and re-explains the same context, that's a skill asking to be born. Signals: a recurring chore with stable steps, a house style that must survive stateless sessions, a checklist someone keeps retyping.
How to mint one (the seat runs this; the human gates it)
- Fetch the source: read the skill-creator skill from github.com/anthropics/skills (a Code session can attach or fetch the repo; a chat seat can read it through its GitHub connector).
- Follow it — it covers the parts homemade skills usually get wrong: a
descriptionthat actually triggers at the right moments (and only those), progressive disclosure so the skill doesn't flood context, concrete examples over abstractions, and testing the skill against real prompts before trusting it. - House requirements for any skill minted here:
- It lands in
skills/<name>/by PR, like every change (RULE 14). - Its description says explicitly which seat runs it and which room it's invoked in — this factory's humans should never have to guess where a skill lives (hard-learned).
- If it's for a chat-family seat, the PR description reminds the human to upload it in Settings → Skills.
- It lands in
- Journal the mint — what gap the skill closes, so a future retro can check whether it worked.
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 First seen · 23 lines · 97 tokens per session scan A f49e04cb5482
skill-creator (pointer) is a skill published in the GitHub repository squidbay/factory (5 stars, last pushed 12d ago), licensed MIT. It adds 97 tokens to every session and 563 once invoked, about $0.0005 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
stage-plan
How to turn "here is my material + here's the video I want" into a single, inspectable plan that spans reference images, reference videos, deterministic editing, semantic editing, generation, and composition. The output is project/plan.json — a cross-modal Edit Decision List (EDL) — which the assembler walks.…
frontend-design
Use this as the design-lead layer for VideoStudio COMPOSE work. It shapes project/composition/composition-manifest.json::artdirection and the model-authored HTML/SVG motion graphics in index.html.
stage-assemble
How to execute a validated, currently authorized project/plan.json into one finished file. Start/resume with production.status; if planapprovalcurrent is false, pass the state to gate-control and stop instead of producing or opening child gates. Walk the signed EDL; do not re-plan. Host-neutral: VideoStudio-specific…
ppt-review
Read this skill only after current officereview structural output and relevant rendered images exist. Do not use source code, a tool success flag, or the first-slide preview as a substitute for deck evidence.
ui-design-executor
Use this as UIDesigner's default execution skill. It contains the baseline design, accessibility, HTML, runtime-safety, responsive, taste, and verification rules needed by ordinary UI work. Do not load the separate system, control, taste, color, renderer, or craft skills merely to repeat these baseline rules.
ui-craft-checks
Use this comprehensive gate for formal review, QA, launch handoff, exact-fidelity inspection, high-risk complex UI, or when the fast gate exposes a deeper craft problem. Ordinary single-screen work uses ui-design-executor and its bundled validator without loading this full matrix. Pair standalone artifacts with…