axel-pm/watch-n-learn / skills/watch-and-learn
watch-and-learn
VerifiedWatch a how-to / tutorial / demo video and turn the procedure it teaches into one or more working Claude Code skills. Use when the user wants to "turn this video into a skill", "watch this tutorial and make a skill", "learn how to do X from this video", "build a skill from this screencast/demo/talk", or capture a repeatable workflow shown in a video. Sends an agent to watch the video (local file or URL) with a local, free, no-API toolchain, breaks the procedure down, authors the skill(s), and verifies they actually work.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 7.3 KB |
| references/procedure-spec-template.md | markdown | 1.5 KB |
| references/qualifying.md | markdown | 2.9 KB |
| references/skill-authoring.md | markdown | 2.5 KB |
| references/validation-checklist.md | markdown | 1.8 KB |
| references/watching.md | markdown | 4.6 KB |
| scripts/install-tools.shexecutable | script | 1.7 KB |
| scripts/validate_skill.py | script | 4.4 KB |
| scripts/watch.py | script | 16.3 KB |
references/skill-authoring.md
View on GitHub# skill-authoring.md — how to write a good Claude Code skill
A skill is a folder with a `SKILL.md` (YAML frontmatter + Markdown body), optionally `scripts/`
and `references/`. It's loaded by its `description` and, once triggered, its body instructs the agent.
## Frontmatter
```yaml
---
name: kebab-case-capability # matches the folder name
description: >
What it does + explicit "use when …" triggers (the phrases a user would actually say).
This is the ONLY thing the model sees when deciding to fire the skill — make it earn its trigger.
---
```
- `name`: lowercase, hyphenated, names the **capability** (not the source video).
- `description`: one to three sentences. Lead with the capability, then concrete trigger phrases
("use when the user wants to …", quoting real user language). Vague descriptions = skills that never fire.
## Body — principles
- **Action-oriented.** Tell the agent what to *do*, in order. Imperative voice.
- **Lean.** Keep `SKILL.md` scannable. Push long tables, templates, and background into `references/`
and link to them — the agent reads a reference only when it needs it.
- **Deterministic work → scripts.** If a step is exact and repeatable, ship a script in `scripts/`
and call it, rather than describing it in prose the model might drift from.
- **Self-contained.** Don't assume state from the video (paths, secrets, names). Parameterise.
- **State the guardrails.** Compliance, safety, "don't do X" — inline, briefly.
## The boundary rubric (one skill vs several)
Make it **one** skill when:
- the steps only make sense as a single flow, and
- there's a single trigger a user would say, and
- splitting would create fragments nobody would invoke alone.
Split into **several** when the video teaches capabilities that:
- have distinct triggers ("set up X" vs "debug X"), or
- are independently reusable, or
- would otherwise force one bloated `SKILL.md` with unrelated branches.
When in doubt, prefer **one focused skill** over a sprawling one; you can extract later.
## Structure to aim for
```
<name>/
SKILL.md # triggers + the workflow, lean
scripts/<tool>.py # deterministic steps, each with --help
references/<topic>.md # depth loaded on demand
```
## Smells to avoid
- A `description` with no trigger phrases.
- A `SKILL.md` that's a wall of text with no ordered steps.
- Prose reimplementing something that should be a script.
- Hardcoded machine paths, API keys, or the demo's specific filenames.
- A skill that was never run before being called "done".
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |