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 Elumenotion/GuideAnts --skill talking-headgit clone --depth 1 https://github.com/Elumenotion/GuideAntsWrote 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/elumenotion/guideants/talking-head)<a href="https://agentmods.dev/skills/elumenotion/guideants/talking-head"><img src="https://agentmods.dev/badge/skills/elumenotion/guideants/talking-head/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/elumenotion/guideants/talking-head"><img src="https://agentmods.dev/badge/skills/elumenotion/guideants/talking-head.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00068 | $0.01633 |
| Opus 5.5 | $0.00027 | $0.00653 |
| Sonnet 5.5 | $0.00014 | $0.00327 |
| Haiku 4.5 | $0.00007 | $0.00163 |
Grade C, and why
talking-head scanned grade C with 1 finding 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.
Harvests environment variableshighData exfiltration
Enumerating or grepping the environment for keys collects credentials unrelated to what the mod says it does.
printenv TALKING_HEAD_SKILL_BASE_URL >/dev/null && printenv TALKING_HEAD_SKILL_TOKEN >/dev/null && echo "env ok" || echo "env missing" How it starts
The opening of the file, as written. The whole thing — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Talking-head
This is the only talking-head skill. Avatar + audio + background plate → 1280×720 MP4.
One adapter job: LongCat-Video-Avatar-1.5 at 416×256 → crop 416×234 → CorridorKey @ native → BasicVSR++ 4× on keyed FG → 720p. Plate blur σ=1.5 before CK; FG sharpen 0.
The adapter workflow id is still infinitetalk-i2v-v1. Do not invent a different
workflow name. Generation inside that graph is LongCat-Video-Avatar-1.5 (Whisper-large-v3,
8-step DMD distill).
Sandbox CWD is the notebook output directory. Scripts:
Skills/talking-head/scripts/. Write the MP4 with a bare filename
(-o talking-head.mp4). Never prefix -o with Output/.
Do not invent a POST. scripts/video_tool.py submits: files source / audio /
background; form field parameters is one JSON object. Do not pass
--width --height --steps --cfg --fps unless the user explicitly asks to
deviate. Do not call ComfyUI /free. Do not start a second job while this one runs.
Do not cancel CorridorKey frame 0. V2V is not available.
Environment
Scripts read TALKING_HEAD_SKILL_BASE_URL and TALKING_HEAD_SKILL_TOKEN from the
guide Environment. Do not hardcode or export them inline.
printenv TALKING_HEAD_SKILL_BASE_URL >/dev/null && printenv TALKING_HEAD_SKILL_TOKEN >/dev/null && echo "env ok" || echo "env missing"
If either is missing, stop and ask the user to set them. Never scan the LAN.
Contract: references/parameters.md. Stages and measured times: references/workflows.md.
How long this takes (measured)
A sandbox script call is killed after about 10 minutes. These jobs last much
longer than that. i2v submits and exits. You poll status on later calls.
The adapter sets frames from audio duration at 25 fps (max 7200). Times below
are InfiniteTalk 4-step clips at 10.56s / 264 frames. LongCat-Video-Avatar-1.5
is 8-step distill on a larger DiT; sampling will take longer than this table.
Keep polling until state is completed or failed.
What ships with it
6 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.
- 1mo ago Changed · +74 lines · +22 tokens per session scan A → C 026f8b02ac2c
- 1mo ago First seen · 64 lines · 46 tokens per session scan A d26d73a5143b
talking-head is a skill published in the GitHub repository Elumenotion/GuideAnts (61 stars, last pushed 2d ago), licensed Apache-2.0. It adds 68 tokens to every session and 1,633 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it C with 1 finding (harvests environment variables). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
example-feature-generator
Generate boilerplate for new features following project conventions. Use when user says "create feature", "add feature", or "new feature scaffolding".
thinking-theory-of-constraints
When throughput or latency is pipeline-limited, identify the single binding constraint and exploit, subordinate, elevate, then recheck—ignore non-constraints.
agent-flutter-developer
Specialist subagent: Flutter/Dart cross-platform development, Riverpod state management, and platform channels specialist. Use when building Flutter apps, implementing state management, or working with native platform features. Trigger phrases: Flutter, Dart, Riverpod, Bloc, widget, MaterialApp, GoRouter, platform…
content-repurposer
Adapt content across platforms with tone/format shifting — blog to social, long to short, text to visual outline. Use when repurposing content for different channels, audiences, or formats.
electron-desktop
Desktop application development with Electron for Windows, macOS, and Linux. Use when building cross-platform desktop apps, implementing native OS features, or packaging web apps for desktop.
investment-memo-generator
Investment memo creation combining financial analysis, document generation, and structured templates. Use when creating investment memos, pitch decks, deal summaries, or investment committee materials.