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 blitzreels/agent-skills --skill blitzreels-clippinggit clone --depth 1 https://github.com/blitzreels/agent-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/blitzreels/agent-skills/blitzreels-clipping)<a href="https://agentmods.dev/skills/blitzreels/agent-skills/blitzreels-clipping"><img src="https://agentmods.dev/badge/skills/blitzreels/agent-skills/blitzreels-clipping/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/blitzreels/agent-skills/blitzreels-clipping"><img src="https://agentmods.dev/badge/skills/blitzreels/agent-skills/blitzreels-clipping.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 | $0.00049 | $0.00561 |
| Opus 5 | $0.00024 | $0.00280 |
| Sonnet 5 | $0.00010 | $0.00112 |
| Haiku 4.5 | $0.00005 | $0.00056 |
Grade A, and why
blitzreels-clipping 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 — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
BlitzReels Clipping
Use the managed clip state machine. It owns ingest, transcription, selection, layout, captions, QA, repair, and export sequencing.
Steps
- Resolve the source as a supported public video URL or workspace asset ID and resolve selection intent. Use a provided time range or suggestion; use automatic best selection only when choice is delegated. Complete when source and selection mode are explicit.
- Choose the available surface. Prefer the bundled MCP
clips_createtool; otherwise discover the current contract withblitzreels agent-context --command "clips create" --jsonor the OpenAPI/clipsoperations. Complete when required inputs, mutation effects, and state fields are known. - Create with a caller-generated idempotency key.
Complete when the response contains a clip ID, project ID, state, and
next_action. - For MCP batches, follow
batch.pollingwithclips_get. For managed API clips, follownext_action: poll, reselect, repair, export, or stop. Complete each transition only when the next response advances or reaches a terminal state. - When the user retained selection choice, present available alternatives with title, hook, range, duration, and score; reselect the approved ID or range. Complete when the chosen selection is persisted.
- Validate layout intent. Tutorial layout is for flattened screen-share plus facecam footage. Interview and podcast footage use automatic, split, or focus layouts. Complete when applied layout and content type agree.
- Review QA previews and warnings; run only the repair action offered by current state. Complete when QA passes or the state reports a specific unresolved block.
- Start a paid export only after user approval and an export-ready state. Poll to terminal and verify output duration, aspect ratio, captions, framing, and download location.
Manual branch
Use low-level project media, transcript, timeline, caption, and preview operations only when managed state reports a specific block requiring manual control.
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 · 50 lines · 49 tokens per session scan A 7e340c8440b6
blitzreels-clipping is a skill published in the GitHub repository blitzreels/agent-skills (3 stars, last pushed 10d ago), licensed MIT. It adds 49 tokens to every session and 561 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
authoring-video-templates
Use when creating or editing a video template JSON (the template descriptor), adding or changing sections/filters/maps/variables/transitions/looks/motion/audio/layers, or debugging template validation errors in ffmpeg-video-composer.
core-architecture-patterns
Use when adding a segment type, platform adapter, editor manager, core service, or descriptor effect (look/grade/motion/section-audio field) in packages/ffmpeg-video-composer, or when wiring new dependencies into the tsyringe container.
ondevice-ffmpeg-engine
Use when building, modifying, or consuming the on-device FFmpeg engine — the Rust crate packages/ffmpeg-engine, the leclap-ffmpeg Expo native module, the run/probe/version/cancel API, the uniffi bindings, build-engine.sh, or the FFmpeg-from-source toolchain in scripts/ffmpeg.
cross-platform-ffmpeg
Use when working with FFmpeg across Node/Static/WASM, the PlatformBridge, the FFmpeg detection/fallback chain, or browser/React Native runtime constraints in ffmpeg-video-composer.
kling-studio
Full-featured Kling 3.0 Omni video generation skill. Covers text-to-video, image-to-video, video editing (base mode), video reference (feature mode), multi-shot generation, and audio-synced video. Includes validated API constraint rules and prompt engineering guide.
seedance-prompt
This skill should be used when the user asks to write, improve, translate, compress, or debug a Seedance 2.0 video prompt; mentions T2V, I2V, V2V, R2V, camera direction, prompt quality, or provides reference assets for a production-ready prompt.