yasyf/cc-skills

## Get started
text
/plugin marketplace add yasyf/cc-skills
That one add registers the marketplace as skills. Pick plugins from the /plugin browser, or install one directly:
text
/plugin install slop-cop@skills
Driving with an agent? Paste this:
text
/plugin marketplace add yasyf/cc-skills
/plugin install slop-cop@skills
/plugin install repo-bootstrap@skills
/plugin install cli-demo@skills
## Use cases
### Catch LLM prose tells before they ship
You stopped noticing the hedge stacks, em-dash pileups, and delves in your own READMEs and PR descriptions months ago. With slop-cop installed, ask:
text
Run slop-cop on README.md
The skill runs the prebuilt slop-cop binary and reports each violation with its line and rule. A SessionStart hook fetches the binary, so no Go toolchain is required. Report-only by default; it rewrites only when you ask.
### Scaffold a Python, Go, or Swift repo with docs, hooks, and releases wired in
Every new repo means re-deriving the same conventions by hand, from agent docs and Claude Code settings to lint config and a release pipeline. With repo-bootstrap installed, ask:
text
Bootstrap a new Python CLI repo called trimmy, with docs and PyPI releases
One pass renders AGENTS.md, Claude Code settings, guard hooks, brand images, and a uv Python layer with Click, pytest, ruff, and ty, plus a Great Docs site and tag-driven PyPI trusted publishing wired in. The Go and Swift layers ship Homebrew releases the same way.
### Record an animated SVG terminal demo of your CLI
Hand-typed demo GIFs lie, and they go stale the first time your output changes. With cli-demo installed, ask:
text
Record an animated terminal demo of trimmy --help for the README
The skill writes a .tape script, renders it with evp, inspects still keyframes, and refines in a loop. On Linux x8664 the render runs natively; on macOS/ARM it runs under Docker linux/amd64. The tape is committed, so the demo regenerates whenever the output drifts.
## Plugins
The table below is the full catalog; the SKILL.md inside each plugin dir carries prerequisites and the full flow.
| Plugin | What you get |
| slop-cop | Catches LLM prose tells in any file, from filler intensifiers to negation pivots. A SessionStart hook fetches the prebuilt binary. |
| codex | A second opinion from OpenAI's Codex CLI when you're stuck, plus image generation via its $imagegen skill. Needs codex on PATH. On Linux amd64, owner wakes need codex-ask supervise running in the foreground; any process of the same user can reach its daemon, so run it only on a single-user machine. |
| repo-bootstrap | A new repo with agent docs, guard hooks, brand images, and opinionated Python, Go, or Swift layers, scaffolded in one pass. |
| llm-prompts | Research-backed prompt-writing guidance, refreshed with current per-provider model behaviors. |
| writing-docs | Diataxis modes, a technical-builder voice, and the README skeleton this file follows. |
| gen-image | Mascot logos, README banners, social cards, and illustrations, compressed locally to under 1 MiB. Needs an OPENAIAPIKEY. |
| gh-profile | A fancy GitHub profile README built from your real repos and activity, kept fresh by cron Actions. |