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 sutchan/Agent-Skills-Hub --skill opus-clipgit clone --depth 1 https://github.com/sutchan/Agent-Skills-HubWrote 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/sutchan/agent-skills-hub/opus-clip)<a href="https://agentmods.dev/skills/sutchan/agent-skills-hub/opus-clip"><img src="https://agentmods.dev/badge/skills/sutchan/agent-skills-hub/opus-clip/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/sutchan/agent-skills-hub/opus-clip"><img src="https://agentmods.dev/badge/skills/sutchan/agent-skills-hub/opus-clip.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 | $0.00088 | $0.02017 |
| Opus 5 | $0.00044 | $0.01009 |
| Sonnet 5 | $0.00018 | $0.00403 |
| Haiku 4.5 | $0.00009 | $0.00202 |
Grade A, and why
opus-clip 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 8d 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.
This is a copy
92% identical to opus-clip — 18 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
opus-clip
The clip-pipeline tool skill — choose the source, let AI find candidates, inspect and rank honestly, polish
the top few, ship on a drip. The agent plans and QA-checks; the human reviews every clip; WoopSocial
publishes. (Ships with tools/integrations/opus-clip.md; pairs with the captions-and-clipping craft skill.)
The POV: a first-pass clip finder, not a one-click viral machine
OpusClip genuinely compresses 3–5 hours of manual clipping into minutes of processing plus review — its ClipAnything engine is multimodal (visuals + audio sentiment + facial expressions), so it finds moments even in wordless content, and natural-language moment search pulls one moment from a 2-hour VOD without scrubbing. But the top-1% operator holds two facts the marketing won't volunteer. (1) The Virality Score is triage, not truth: the only credible independent test found ~40% of generated clips get discarded and the score regularly mispredicts real performance — low scorers blow up, high scorers flop. Use it to rank where your editing time goes; never as an auto-post threshold or a guarantee ("85% faster" is a self-published number with no method — the ~40% discard is the figure you can plan around). (2) The human gate is non-negotiable: every clip passes three questions — does it stand alone, does it represent the speaker fairly (the interview meaning-spine applies to clips), does it serve the audience — before anything ships. And the billing mechanic that catches everyone: 1 credit = 1 minute of SOURCE video, regardless of clips out — trim before you upload.
Read these first
- captions-and-clipping — the general clipping/repurposing craft this tool executes.
- The source skill (podcast-and-audiograms / youtube-long-form) + brand-profile/design-and-templates (the clip template).
The framework: CLIPS
(Depth: references/the-clips-framework.md.)
- C — Choose the source: conversational long-form clips best (~8–12 candidates/hour); non-verbal genres work via multimodal analysis with higher discard — test on Free first; clean audio, no burned-in subs, trim before upload (credits bill on source minutes).
- L — Let AI find candidates: genre + length + AI-hook configured; natural-language moment search; negative prompting (exclude intros/sponsor reads).
- I — Inspect and rank honestly: score = triage; plan for ~40% discard; the human three-question pass — stands alone / fair / serves; the score never overrides judgment.
- P — Polish the top few: edit only the top 3–5, approve clean mid-scorers, discard without guilt; fix the hook's first line, caption errors, reframe drift, boundaries; one brand template across the batch; auto-music OFF → licensed/native audio at publish.
- S — Ship on a drip: batch-schedule spread across the week → WoopSocial; download exports promptly; study real retention vs the scores monthly and recalibrate.
What ships with it
5 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.
- 8d ago First seen · 103 lines · 88 tokens per session scan A 292b183dc5a0
opus-clip is a skill published in the GitHub repository sutchan/Agent-Skills-Hub (2 stars, last pushed yesterday), licensed MIT. It adds 88 tokens to every session and 2,017 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. It is 92% identical to opus-clip, differing in 18 lines, and is treated as a copy.
Other skills, from other repositories
writing-skills
Use when creating new skills, editing existing skills, or verifying skills work before deployment.
receiving-code-review
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation.
writing-plans
Use when you have a spec or requirements for a multi-step task, before touching code.
skill-authoring
Author SKILL.md skills: frontmatter, validator limits, structure.
skill-creator
Create, improve, evaluate, benchmark skills. Use when authoring a new skill, updating an existing one, running evals, or optimizing a skill's description for triggering. Don't use for invoking skills, writing prose, or scaffolding Python projects.
iflytek-hyper-tts
A text-to-speech tool that turns written text into an MP3 recording. It can use an authorized voice and adjust speaking speed, volume, and pitch.