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 calesthio/generative-media-skills --skill reference-media-analysisgit clone --depth 1 https://github.com/calesthio/generative-media-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/calesthio/generative-media-skills/reference-media-analysis)<a href="https://agentmods.dev/skills/calesthio/generative-media-skills/reference-media-analysis"><img src="https://agentmods.dev/badge/skills/calesthio/generative-media-skills/reference-media-analysis/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/calesthio/generative-media-skills/reference-media-analysis"><img src="https://agentmods.dev/badge/skills/calesthio/generative-media-skills/reference-media-analysis.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 59 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00087 | $0.06060 |
| Opus 5 | $0.00044 | $0.03030 |
| Sonnet 5 | $0.00017 | $0.01212 |
| Haiku 4.5 | $0.00009 | $0.00606 |
Grade A, and why
reference-media-analysis 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 — 438 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Reference Media Analysis
Use reference media to understand what works, not to reproduce protected expression, private identity, or misleading claims. Treat each reference as evidence for production decisions: extract transferable intent, structure, craft, constraints, and quality bars; avoid copying distinctive expression unless the client has explicit rights and approvals for that exact use.
This skill is not legal advice. It gives production triage, documentation, and escalation triggers. When rights, likeness, voice, trademark, privacy, endorsement, or jurisdictional questions materially affect release, pause and ask for client/legal approval.
The governing rule: extract function, not identity
For every reference, separate:
- Documented facts: what the file visibly or audibly contains; metadata; rights and consent statements supplied by the client.
- Production inferences: why the reference works; what creative function each element serves.
- Heuristics: safe ways to adapt the function into a new work.
- Prohibited copying risks: distinctive expression, private identity, voice/likeness, trademarks, product claims, or platform-regulated synthetic media disclosures.
Convert "make it like this" into "make a new work with the same production function." Examples:
- Do extract "cold open in 0.8 seconds, problem stated before brand reveal, high-contrast captions."
- Do not copy the exact edit, catchphrase, character, layout, choreography, melody, camera sequence, voice, celebrity likeness, or trade dress.
- Do extract "warm, credible founder-read with relaxed pacing."
- Do not clone a real speaker's voice or imply their endorsement without documented consent.
Start with a clearance and risk gate
Before detailed creative analysis, build a reference ledger. If inputs are incomplete, continue with visible/audio analysis but mark missing permissions as blockers for production use.
Minimum ledger fields:
reference_id: ref_001
asset_type: image | video | audio | storyboard | mood_board | brand_asset | product_shot | prior_cut | performance
source_location: local path, URL, client upload, DAM ID, or physical source note
source_owner_or_provider: known owner, platform, agency, photographer, performer, unknown
provided_by: client | internal team | public web | user upload | stock provider | other
permission_basis: owned | licensed | client-approved | public-domain claim | fair-use claim | unknown
license_or_approval_evidence: contract ID, email, rights memo, release, invoice, stock license, none
allowed_uses: analysis only | prompt reference | direct incorporation | brand/product fidelity | internal review only
restricted_uses: no training, no public release, no likeness, no voice, no logo, no minors, no paid ads, etc.
people_present: none | private person | public figure | employee | actor | minor | unknown
likeness_or_voice_risk: none | low | medium | high | blocker
trademark_or_trade_dress_risk: none | low | medium | high | blocker
privacy_or_sensitive_data: none | faces | locations | medical | financial | credentials | minors | other
synthetic_or_manipulated_status: captured | edited | AI-generated | mixed | unknown
provenance_checked: C2PA/Content Credentials, IPTC, EXIF/XMP, file history, none
client_approval_status: approved | pending | rejected | not requested
analysis_date: YYYY-MM-DD
notes:
What ships with it
1 file 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.
- 10d ago First seen · 438 lines · 87 tokens per session scan A 958de1fb10f1
reference-media-analysis is a skill published in the GitHub repository calesthio/generative-media-skills (170 stars, last pushed 1mo ago), licensed MIT. It adds 87 tokens to every session and 6,060 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 skills, from other repositories
feature-demo-recording
Record a demo video of a web feature from a real browser. Two modes -- a NARRATED film where measured voiceover drives the timeline (designed slides, subtitles, punch-in camera, rendered from an HTML timeline), and a SILENT evidence clip for a PR or a QA pass. Use when the user asks to record a video, demo, or screen…
image-authoring
Author images and diagrams as code — SVG, Pillow, Excalidraw, mermaid. Load when asked to draw, illustrate, or make an image, icon, logo, poster, or diagram.
bento-slides
Create and edit Bento presentations — self-contained .bento.html decks whose document is JSON. Use whenever the user wants a slide deck or presentation: from scratch, from source material, or by improving an existing file.
pptx-maker
Generate or restyle a PowerPoint deck. Use when the user wants to create or edit a .pptx presentation, build slides from text or a URL, or design a reusable slide style.
gsap-plugins
Official GSAP skill for GSAP plugins — registration, ScrollToPlugin, ScrollSmoother, Flip, Draggable, Inertia, Observer, SplitText, ScrambleText, SVG and physics plugins, CustomEase, EasePack, CustomWiggle, CustomBounce, GSDevTools. Use when the user asks about a GSAP plugin, scroll-to, flip animations, draggable, SVG…
seedance-2-0
Generate cinematic clips with ByteDance Seedance 2.0 — the preferred premium video model in OpenMontage when a paid gateway is configured. Use when: (1) producing trailers, teasers, hype edits, or premium cinematic clips, (2) needing native synchronized audio (speech, SFX, ambience) in a single pass, (3) needing…