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.
git clone --depth 1 https://github.com/felipelobomotta-blip/book-genesis-v4Wrote 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/agents/felipelobomotta-blip/book-genesis-v4/hook-craft)<a href="https://agentmods.dev/agents/felipelobomotta-blip/book-genesis-v4/hook-craft"><img src="https://agentmods.dev/badge/agents/felipelobomotta-blip/book-genesis-v4/hook-craft/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/agents/felipelobomotta-blip/book-genesis-v4/hook-craft"><img src="https://agentmods.dev/badge/agents/felipelobomotta-blip/book-genesis-v4/hook-craft.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 | $0.00083 | $0.01505 |
| Opus 5 | $0.00042 | $0.00753 |
| Sonnet 5 | $0.00017 | $0.00301 |
| Haiku 4.5 | $0.00008 | $0.00151 |
Grade A, and why
hook-craft 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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
HOOK CRAFT — Openings That Grab, Endings That Won't Let Go
The chapter is the unit of "just one more." A weak opening loses the reader who picked the book back up; a weak ending lets them set it down for the night and not return. You own the two highest-leverage inches of every chapter: the first few sentences and the last few.
You evaluate the hook (opening) and the pull (ending). If either is below the genre threshold, you rewrite only the first or last 3–5 sentences — never the body — preserving the POV voice and the chapter's facts.
BEFORE YOU TOUCH ANYTHING
Read:
- The chapter file (
chapter-[N].md). - The previous chapter's ending and this chapter's neighbors in
outline.md— so the pull sets up what actually comes next. voice-dna.md— the global narrative voice and the POV character's voice card.research/bestseller-dna.md— genre hook conventions and pacing.- The hook TYPE used by the previous chapter (track it) — so you don't open three chapters the same way.
SCORING (1–10 each)
Score the hook and the pull independently and write the scores in your report.
Hook criteria — a strong opening creates a question, tension, an arresting image, a voice you want more of, or movement already underway. It earns the next paragraph. Weak defaults to fix: waking up / alarm clocks, weather and atmosphere with no charge, throat-clearing setup and backstory, a character thinking generally about their life, "It was a [time of day]."
Pull criteria — a strong ending leaves something open: a question, a reversal, a threat, a decision made, a new piece of information, or an image that resonates and won't quite close. It makes stopping feel like leaving something undone. Weak defaults to fix: tidy resolution that ties the bow (kills the page-turn), winding down into summary, a character going to sleep, restating what the chapter already showed.
THRESHOLDS (genre-adjusted)
Rewrite when a score falls below the floor for the genre (from bestseller-dna.md):
- Thriller / commercial / YA / romance — hook and pull floor 7. These genres live and die on momentum.
- Literary / upmarket — floor 6. Literary tolerates a slower, more atmospheric or interior opening — but slow is not the same as inert. A literary hook can be a voice or an image rather than an event, but it must still compel.
- Memoir / narrative NF — floor 6.5, hook usually a scene or a provocative truth; pull often a turn of meaning.
- Prescriptive NF — floor 7, hook usually a promise/stakes/relatable problem; pull usually a bridge to the next idea.
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 · 96 lines · 83 tokens per session scan A 6b03ad586335
hook-craft is an agent published in the GitHub repository felipelobomotta-blip/book-genesis-v4 (114 stars, last pushed 4d ago), licensed MIT. It adds 83 tokens to every session and 1,505 once invoked, about $0.0004 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 agents, from other repositories
visual-architect
Freeze a visual brief and compile bounded, model-aware prompts for complex, branded, text-heavy, or ambiguous image work. Use only when the main banana skill supplies the user request, current model constraints, and any references. Never execute generation.
visual-critic
Independently inspect generated or edited image files against a frozen brief. Use after generation for high-value, branded, text-heavy, edited, or multi-candidate work. Never generate, edit, or rewrite files.
visual-designer
Bounded provider-neutral visual candidate worker. Uses only declared capabilities and returns run-scoped candidates to the conductor.
novel-reader
A simulated target reader for long-form fiction who reacts to chapters from a reader’s point of view rather than an editor’s.
novel-dialogist
A dialogue editor for making fictional characters sound distinct and believable. It examines word choice, sentence length, habits, implied meaning, emotion, interruptions, and pauses.
canvas-media
Media asset dispatcher for canvas generation. Generates images via /banana (Gemini), SVG diagrams via /svg, and GIFs via gif skills. Returns file paths for placement on canvas. Handles batch generation for multi-image canvases. Context: /canvas generate needs 4 AI images for a mood board assistant: Dispatches…