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 Glad-Labs/poindexter --skill videogit clone --depth 1 https://github.com/Glad-Labs/poindexterWrote 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/glad-labs/poindexter/video)<a href="https://agentmods.dev/skills/glad-labs/poindexter/video"><img src="https://agentmods.dev/badge/skills/glad-labs/poindexter/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/glad-labs/poindexter/video"><img src="https://agentmods.dev/badge/skills/glad-labs/poindexter/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 | $0.00058 | $0.00839 |
| Opus 5 | $0.00029 | $0.00419 |
| Sonnet 5 | $0.00012 | $0.00168 |
| Haiku 4.5 | $0.00006 | $0.00084 |
Grade A, and why
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 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Video skill
One prompt the pipeline uses to turn a finished article into a spoken
short-form video narration. The architect routes on the description
above; UnifiedPromptManager resolves the template by key (Langfuse
override still wins over the body below).
Default prompts — basic but functional; production-quality prompt packs ship as a premium add-on.
video.short_form_narration
Write a 60-second video narration (about 150 words) summarizing this article.
RULES:
- Start with a compelling hook that grabs attention in the first 5 seconds
- Cover the 2-3 most important takeaways
- End with a call to action inviting viewers to read the full article at {site_name}
- Conversational, energetic tone — this is for TikTok/YouTube Shorts
- No URLs, no markdown, no special characters
- Write ONLY the narration text, nothing else
ARTICLE TITLE: {title}
ARTICLE CONTENT:
{content}
NARRATION:
The long-form prompt's {target_seconds} / {target_words} placeholders are
substituted from video_long_target_seconds (words = seconds × 2.5 WPS) — the
same one-canonical-target pattern as the short lane, so the narration ask, the
director's visual plan, and the runaway-trim ceiling can never disagree.
(Prose must sit OUTSIDE the ## <key> → fence pair: extract_section matches
a fence immediately after the heading, so a paragraph between them makes the
key unresolvable and the stage falls back to its in-code default.)
video.long_form_narration
Write a voiceover narration script for a long-form video about the article below.
The narration is spoken aloud and must stand on its own as audio. Write it for the ear: explain the subject directly to the listener. Do not refer to any accompanying imagery — the supporting footage is generic and will not match specific visual references, so keep every line meaningful with the eyes closed.
- Aim for a ~{target_seconds}-second narration (about {target_words} words of spoken prose).
- COLD OPEN: start mid-thought on the article's strongest concrete fact or tension. Never open with a greeting or a scene-setting frame — no "Welcome", "In today's", "Let's explore", "Imagine", "deep dive".
- Close on the article's final insight in one natural sentence. Never "In conclusion", "In summary", "To wrap up". Do NOT add a like/subscribe call-to-action — that is appended separately.
- Keep every number, dollar figure, and statistic exactly as the article states it — the numbers are the substance.
- Banned words and phrases: delve, tapestry, testament, game-changer, revolutionize.
- Plain spoken prose. Commas and periods, not semicolons. No headings, no stage directions, no emojis, no markdown.
TITLE: {title}
ARTICLE:
{content}
NARRATION:
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 · 78 lines · 58 tokens per session scan A e78b265542c3
video is a skill published in the GitHub repository Glad-Labs/poindexter (5 stars, last pushed today), licensed Apache-2.0. It adds 58 tokens to every session and 839 once invoked, about $0.0003 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
content-pipeline
A Chinese-language workflow for turning collected source material into articles and versions for several publishing platforms, including formatted previews, cover images, social posts, podcasts, and videos.
openai-image-gen
Batch-generate images via OpenAI Images API. Random prompt sampler + index.html gallery.
nano-banana-pro
Generate or edit images via Gemini 3 Pro Image (Nano Banana Pro).
video-frames
Extract frames or short clips from videos using ffmpeg.
sherpa-onnx-tts
Local text-to-speech via sherpa-onnx (offline, no cloud).
sag
ElevenLabs text-to-speech with mac-style say UX.