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/whystrohm/shotkit/brand-lock-extractornpx skills add whystrohm/shotkit --skill brand-lock-extractorgit clone --depth 1 https://github.com/whystrohm/shotkitWhat 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 | $0.00126 | $0.01816 |
| Opus 5 | $0.00063 | $0.00908 |
| Sonnet 5 | $0.00025 | $0.00363 |
| Haiku 4.5 | $0.00013 | $0.00182 |
Grade A, and why
brand-lock-extractor 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.
How it starts
The opening of the file, as written. The whole thing — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Brand Lock Extractor
The blank brand-packs/_template.md is the single biggest point of friction in shotkit. Nobody wants to hand-author nine sections of brand parameters before they can produce a single storyboard. This skill removes that wall: hand it what you already have, get a filled brand-lock back.
The output is a brand-lock.md in the exact format tools/validate_brand_lock.py validates and every other skill consumes. Same file the pipeline reads, produced from your assets instead of from scratch.
This skill extracts. It does not invent. Every value is sampled from a real asset or flagged as an estimate the user must confirm. A confident-sounding wrong hex is worse than a flagged guess.
When to use
Trigger when the user:
- Wants to onboard a brand into shotkit and has assets (a site, a brand book, screenshots)
- Says "extract my brand", "build a brand lock", "make a brand pack from my site/PDF"
- Hands over a URL, a PDF, image files, or a written brand description and asks for a brand-lock
- Has a brand-lock that is half-filled and wants the gaps extracted from assets
If the user has nothing but a vague idea (no assets, no description), they are not extracting, they are authoring. Point them at brand-packs/_template.md and help them fill it directly.
Inputs
You can work from any one of these. More sources is a better extraction.
| Source | How you read it | Best for |
|---|---|---|
| Website URL | WebFetch homepage + about + one more page |
Voice, positioning, palette, type |
| Brand book PDF | Read the PDF |
Palette (exact hex), type, motion rules |
| Screenshots / image files | Read the images |
Palette (sample pixels), mood, layout |
| Written description | Use directly | Identity, archetype, voice posture |
At minimum you need one source. If the user offers none, ask for the one they have: "What can I work from, a website URL, a brand book PDF, screenshots, or a short description?" One question.
Workflow
Step 1. Gather the source material
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 123 lines · 126 tokens per session scan A b6517ce9f5f0
brand-lock-extractor is a skill published in the GitHub repository whystrohm/shotkit (18 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 126 tokens to every session and 1,816 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
clinpgx-database
Access ClinPGx pharmacogenomics data (successor to PharmGKB). Query gene-drug interactions, CPIC guidelines, allele functions, for precision medicine and genotype-guided dosing decisions.
embedding-strategies
Select and optimize embedding models for semantic search and RAG applications. Use when choosing embedding models, implementing chunking strategies, or optimizing embedding quality for specific domains.
datavis
Comprehensive data visualization toolkit for creating beautiful, mathematically elegant visualizations with D3.js, Chart.js, and custom SVG. Use when (1) building interactive data visualizations, (2) designing color palettes for charts, (3) choosing scales and visual encodings, (4) creating data pipelines from…
docx-comment-reply
Reply to comments (批注) in Word .docx/.doc files: extract comment context, draft replies, write threaded replies back, and validate OOXML.
vibe-research
Deep research and market validation for app ideas. Use when starting a new project, validating an idea, or when the user says "research my idea", "validate my app", or "help me start a new project".
detecting-data-anomalies
Investigate outliers, rare events, spikes, and suspicious records in datasets. Use as an explicit anomaly-analysis helper when you want concrete anomaly-detection workflow guidance, not generic data validation or end-to-end ML ownership.