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 sunshineLixun/video-publish-skill --skill ego-publishergit clone --depth 1 https://github.com/sunshineLixun/video-publish-skillWrote 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/sunshinelixun/video-publish-skill/ego-publisher)<a href="https://agentmods.dev/skills/sunshinelixun/video-publish-skill/ego-publisher"><img src="https://agentmods.dev/badge/skills/sunshinelixun/video-publish-skill/ego-publisher/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/sunshinelixun/video-publish-skill/ego-publisher"><img src="https://agentmods.dev/badge/skills/sunshinelixun/video-publish-skill/ego-publisher.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 | $0.00080 | $0.06102 |
| Opus 5 | $0.00040 | $0.03051 |
| Sonnet 5 | $0.00016 | $0.01220 |
| Haiku 4.5 | $0.00008 | $0.00610 |
Grade A, and why
video-publisher 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 12d 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 — 247 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Video Publisher
Prepare one confirmed video package and drive selected creator platforms to a verified draft state. Use Ego Lite for all live creator-page work.
Configuration And Onboarding
At the start of every invocation, before inspecting a video or opening a browser, run:
node scripts/config.mjs status
If onboardingRequired is true, stop the publishing flow and onboard the user. Ask first which supported creator platforms the user actually has; require at least one and never assume all four. Then ask which of those available platforms should run by default, proposing all available platforms as the default subset. Ask for Douyin topics only when Douyin is available and Bilibili automatic tags only when Bilibili is available. Collect the source directory, shared copy/tag preferences, and whether every video may truthfully be declared original; keep concurrency 4/4 and platform cover as proposed defaults unless the user changes them. Summarize the choices before writing. Save available accounts with repeatable --available-platform flags and defaults with repeatable --platform flags, run validate, and continue only when onboardingRequired is false.
Configuration is per user at $XDG_CONFIG_HOME/video-publisher/config.json, or $HOME/.config/video-publisher/config.json. VIDEO_PUBLISHER_CONFIG overrides the path. Never put a user's configuration inside the shareable Skill folder.
An explicit current request overrides the package; explicit package fields override configuration defaults. A current request may select any configured available platform, but it cannot silently add an unavailable platform: update onboarding and confirm that account first. The onboarding configuration may persist the user's truthful standing originality policy and declared platform availability, but never cookies, credentials, video-specific paths, or final-publish authorization. Read references/configuration.md for the schema and onboarding command.
What ships with it
23 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.
- LICENSE.upstream 1.0 KB
- scripts/check-package.mjs 1.4 KB runs code
- scripts/config.mjs 5.5 KB runs code
- scripts/find-video.mjs 4.6 KB runs code
- scripts/lib/config.mjs 8.5 KB runs code
- scripts/lib/content-package.mjs 8.7 KB runs code
- scripts/lib/media.mjs 4.3 KB runs code
- scripts/run-safe-platforms.sh 385 B runs code
- scripts/v2/ego/core.mjs 13 KB runs code
- scripts/v2/lib/identity.mjs 1.4 KB runs code
- scripts/v2/lib/job-lock.mjs 2.4 KB runs code
- scripts/v2/lib/job-store.mjs 5.1 KB runs code
- scripts/v2/lib/model.mjs 5.0 KB runs code
- scripts/v2/lib/platform-lock.mjs 1.9 KB runs code
- scripts/v2/lib/result-line.mjs 617 B runs code
- scripts/v2/lib/scheduler.mjs 668 B runs code
- scripts/v2/platforms/bilibili.mjs 25 KB runs code
- scripts/v2/platforms/dispatch.mjs 437 B runs code
- scripts/v2/platforms/douyin.mjs 44 KB runs code
- scripts/v2/platforms/wechat-channels.mjs 23 KB runs code
- scripts/v2/platforms/xiaohongshu.mjs 26 KB runs code
- scripts/v2/publisher.mjs 16 KB runs code
- scripts/v2/run-platform.mjs 6.0 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.
- 12d ago First seen · 247 lines · 80 tokens per session scan A 76005f9538bf
video-publisher is a skill published in the GitHub repository sunshineLixun/video-publish-skill (22 stars, last pushed 8d ago), licensed Apache-2.0. It adds 80 tokens to every session and 6,102 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
dws
A command-line guide for DingTalk, a workplace platform with chat, documents, calendars, approvals, tasks, and other business tools.
tmeet-skill
A command-line tool for Tencent Meeting, a video-conferencing service. It covers sign-in, meetings, recordings, transcripts, reports, invitations, and controls during a meeting.
tencent-docs
A skill for creating, reading, editing, searching, and managing Tencent Docs, an online document and spreadsheet service.
lark-shared
A shared set of rules for using lark-cli, a command-line tool for accessing Lark, the collaboration platform also known as Feishu. It covers setup, login, identity selection, permissions, JSON output, and authorization errors.
government-writing
A skill for writing formal Chinese government documents, including notices and opinions, in the required official format.
package-author
A packaging guide for turning scripts, skills, or an MCP service into a standard plugin package. MCP is a way for an AI assistant to connect to external tools.