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 galfrevn/reel --skill reel-motiongit clone --depth 1 https://github.com/galfrevn/reelWrote 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/galfrevn/reel/reel-motion)<a href="https://agentmods.dev/skills/galfrevn/reel/reel-motion"><img src="https://agentmods.dev/badge/skills/galfrevn/reel/reel-motion/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/galfrevn/reel/reel-motion"><img src="https://agentmods.dev/badge/skills/galfrevn/reel/reel-motion.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 | $0.00146 | $0.02914 |
| Opus 5.5 | $0.00058 | $0.01166 |
| Sonnet 5.5 | $0.00029 | $0.00583 |
| Haiku 4.5 | $0.00015 | $0.00291 |
Grade C, and why
reel-motion scanned grade C with 2 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 2mo 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
the old tail behind — `rm -rf frames/` before re-exporting, or the Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
const manifest: Manifest = await fetch(staticFile("frames/frames.json")).then((r) => r.json()); How it starts
The opening of the file, as written. The whole thing — 269 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Terminal footage in a motion-graphics pipeline
reel edits terminal recordings — it cuts, speed-ramps, zooms, annotates and scores them. It does not animate logos, build kinetic type, or composite product shots, and it should not try to. When a demo needs to become a launch film, reel produces the footage and something else composes it.
--frames-out is that handoff. It writes a constant-rate PNG sequence and a
frames.json manifest describing the edit — every marker, caption, note,
card, zoom and speed ramp, in output seconds and in frame numbers.
That manifest is the reason this handoff is better than exporting a video file. A compositor handed raw footage has to eyeball where the interesting moments are; reel already knows, because it cut the timeline.
Prerequisite: the terminal demo itself. Use the reel skill to record
and edit it first. This skill starts from a .reel file that already
renders the demo you want.
1. Edit before you export
Do not export a raw recording and cut it in the compositor. Trimming dead air, speed-ramping a wait, and zooming the payoff are cheap in reel (sub-second re-renders on a frozen recording) and expensive in a timeline. Get the demo right as a GIF or WebM first, then export.
Two settings matter for this destination, both in the .reel:
[output]
fps = 30 # or 60 — match the composition exactly
size = "1920x1080" # render at final size; never upscale terminal text
Upscaled monospace looks soft and instantly reads as a screen recording. Rendering at the target size lets reel solve the font size to fit, so the text is crisp at 100%.
If the terminal is going in a framed card rather than full-bleed, render at the card's pixel size, not the canvas's.
2. Export the sequence
reel render demo.reel --frames-out frames/
Writes frames/0000.png, 0001.png, … at the configured fps, plus
frames/frames.json. When the .reel configures [audio], it also writes
frames/audio.wav — reel's procedural keystroke track, already in sync with
the frames beside it.
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.
- 2mo ago First seen · 269 lines · 146 tokens per session scan C fa6d985a00e7
reel-motion is a skill published in the GitHub repository galfrevn/reel (13 stars, last pushed 2mo ago), licensed MIT. It adds 146 tokens to every session and 2,914 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it C with 2 findings (recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
reel-motion compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| reel-motiongalfrevn/reel | Skill | 13 | 2mo ago | MIT | C | 146 | Claude Code, Codex |
| asciinemavanducng/skills | Skill | 8 | yesterday | MIT | A | 137 | Claude Code |
| verifyzenbu-labs/terminal-browser | Skill | 4.5k | yesterday | MIT | A | 37 | Claude Code |
| ast-grepvinhnx/VTCode | Skill | 898 | today | Apache-2.0 | A | 24 | Claude Code, Codex |
| agdocumputun/agterm | Skill | 730 | today | MIT | A | 227 | Claude Code |
| skill-installervinhnx/VTCode | Skill | 898 | today | Apache-2.0 | A | 52 | Claude Code, Codex |
| cronnumputun/agterm | Skill | 730 | today | MIT | A | 119 | Claude Code, Codex |