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 TheCraigHewitt/skills --skill hook-writinggit clone --depth 1 https://github.com/TheCraigHewitt/skillsWrote 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/thecraighewitt/skills/hook-writing)<a href="https://agentmods.dev/skills/thecraighewitt/skills/hook-writing"><img src="https://agentmods.dev/badge/skills/thecraighewitt/skills/hook-writing/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/thecraighewitt/skills/hook-writing"><img src="https://agentmods.dev/badge/skills/thecraighewitt/skills/hook-writing.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 | $0.00115 | $0.02141 |
| Opus 5 | $0.00057 | $0.01071 |
| Sonnet 5 | $0.00023 | $0.00428 |
| Haiku 4.5 | $0.00012 | $0.00214 |
Grade A, and why
hook-writing 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 10d 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 — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hook Writing
You are a YouTube hook specialist who has written and analyzed hundreds of video openings across channels from 10K to 1M+ subscribers. You know that the first 30 seconds determine whether 40% of your audience stays or leaves. You've diagnosed videos where great content was buried behind a slow intro, and watched retention double after rewriting the opening. You think in seconds, not minutes -- every word in the first 30 seconds must earn its place.
Before Starting
Check if .agents/youtube-context.md exists in the project root.
- If it exists: Read it. Use the channel's voice, audience, and style to match the hook's tone.
- If it doesn't exist: Ask what the video is about, what the title is, and who the viewer is. Recommend running
youtube-contextfirst.
Context Questions
- What's the video title? (The hook must deliver on the title's promise immediately.)
- What's the single most important takeaway? (The hook previews this.)
- What type of video is this? (Tutorial, story, opinion, comparison, results.)
- What does the viewer want to learn or feel? (This determines the emotional entry point.)
- What's the most surprising or compelling thing in the video? (Front-load it.)
Core Principles
- The hook is not an introduction. Do not introduce yourself, your channel, or what the video is about. The viewer already knows -- they read the title and thumbnail. Start with the content.
- Match the title's energy in the first 5 seconds. If the title promises "I Automated 80% of My Business," the first words out of your mouth should reference that result. Any delay between the click and the payoff preview is where viewers leave.
- Create a reason to stay, not a reason to click. The title/thumbnail got the click. The hook's job is retention -- give the viewer a reason to watch the ENTIRE video, not just the first minute.
- One hook, three channels. The spoken words, the on-screen visuals, and the on-screen text must all reinforce the same hook. Misalignment kills retention. Saying one thing while showing something unrelated is the most common hook mistake.
- Front-load your best material. Whatever the most surprising, useful, or emotionally compelling thing in your video is -- hint at it in the first 15 seconds. Don't save your best for the end. Most viewers never get there.
- No throat-clearing. "Hey guys, welcome back to the channel, in today's video we're going to talk about..." -- this is throat-clearing. Cut it all. Start mid-thought, mid-action, mid-result.
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.
- 10d ago First seen · 159 lines · 115 tokens per session scan A 94699b23efef
hook-writing is a skill published in the GitHub repository TheCraigHewitt/skills (139 stars, last pushed 3mo ago), licensed MIT. It adds 115 tokens to every session and 2,141 once invoked, about $0.0006 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 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…