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 benyki/gtm-engine --skill engine-videogit clone --depth 1 https://github.com/benyki/gtm-engineWrote 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/benyki/gtm-engine/engine-video)<a href="https://agentmods.dev/skills/benyki/gtm-engine/engine-video"><img src="https://agentmods.dev/badge/skills/benyki/gtm-engine/engine-video/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/benyki/gtm-engine/engine-video"><img src="https://agentmods.dev/badge/skills/benyki/gtm-engine/engine-video.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 | $0.00081 | $0.05428 |
| Opus 5.5 | $0.00032 | $0.02171 |
| Sonnet 5.5 | $0.00016 | $0.01086 |
| Haiku 4.5 | $0.00008 | $0.00543 |
Grade A, and why
engine-video 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 1mo 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 — 409 lines — stays where its author put it; the contents beside it link to each section on GitHub.
engine-video
Script to rendered 9:16 mp4. The heaviest engine in the repo — it needs disk, keys and patience — so don't pick it as a first engine unless video is genuinely the channel.
This skill runs any engine folder of type video. The default folder is
video/; paths below (templates/, runs/, inputs/) are inside it, while
brand, accounts, keys and reusable footage/fonts/logos live in shared/
(shared/assets/ — any engine can use them).
Several video engines are the normal shape, one per format, each with its own templates, experiments and metric. Name them after the format they run:
| Folder | Format |
|---|---|
video-app/ |
viral product |
video-vibe/ |
viral vibe |
video-info/ |
informative |
All of them are type video — scaffold one with
--merge --engine video-vibe:video, or copy a folder and empty its runs/
and reports/ (history belongs to the original). Running a single format? Keep
the shipped video/ folder and ignore the rest. Render knobs (resolution,
target_seconds, voice id) live in each folder's engine.json under video,
so two video engines can render differently.
Paths in this file: shared/… means the gtm home (~/gtm by default, or
$GTM_HOME); templates/, inputs/, runs/ and reports/ mean the engine
folder you're running, wherever it lives. The scripts resolve both through
~/gtm/engines.json, so read them as names rather than literal paths.
Three formats, by default
Decide which one you're making before anything else — it sets the length,
the footage, whether there's a voice, and how loud the music is. Full spec and
the editorial rules for each: references/formats.md.
| Length | Voice | Music | Copy from | |
|---|---|---|---|---|
| Viral product — 4s hook + ~12s of the product working | ~16s | usually none | 50% | examples/viral-app-demo.json |
| Viral vibe — one meditative clip, two lines of text, product named only in the caption | 8–15s | never | 50% | examples/viral-vibe.json |
| Informative — text content (an article, a dataset, a series) turned into something watchable | 30–60s | ElevenLabs | 3% (optional) | examples/informative-vocab.json, examples/informative-recap.json |
What ships with it
25 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.
- assets/fonts/OFL.txt 4.3 KB
- assets/fonts/TikTokSans12pt-Bold.ttf 165 KB
- assets/fonts/TikTokSans12pt-Regular.ttf 165 KB
- examples/informative-recap.json 6.3 KB
- examples/informative-vocab.json 5.1 KB
- examples/viral-app-demo.json 4.5 KB
- examples/viral-vibe.json 3.5 KB
- references/advanced.md 6.2 KB
- references/clip-sourcing.md 6.7 KB
- references/duplicate-safety.md 6.0 KB
- references/ffmpeg-recipes.md 3.0 KB
- references/ffmpeg-text-style.md 9.5 KB
- references/floating-text.md 2.8 KB
- references/footage-pexels.md 1.9 KB
- references/formats.md 9.8 KB
- references/hook-guide.md 12 KB
- references/looks.md 2.7 KB
- references/music.md 3.0 KB
- references/posting-api.md 2.0 KB
- references/posting-options.md 3.3 KB
- references/remotion.md 1.7 KB
- references/structure-plan.md 3.5 KB
- references/voiceover.md 2.3 KB
- scripts/combo_check.py 6.5 KB runs code
- scripts/text_style.py 10 KB runs code
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.
- 1mo ago Changed eab279886261
- 2mo ago First seen · 409 lines · 81 tokens per session scan A 1c9eb106b877
engine-video is a skill published in the GitHub repository benyki/gtm-engine (4 stars, last pushed 6d ago), licensed MIT. It adds 81 tokens to every session and 5,428 once invoked, about $0.0003 per session on Opus 5.5. 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.
engine-video compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| engine-videobenyki/gtm-engine | Skill | 4 | 6d ago | MIT | A | 81 | Claude Code, Codex |
| imagegenopenai/codex | Skill | 134k | today | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | today | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | today | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 314k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |