Borrowing it
Nothing to install: this file belongs to vanillagreencom/kendex. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/vanillagreencom/kendex/main/.agents/skills/youtube-publish/SKILL.mdgit clone --depth 1 https://github.com/vanillagreencom/kendexWrote 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/vanillagreencom/kendex/youtube-publish)<a href="https://agentmods.dev/skills/vanillagreencom/kendex/youtube-publish"><img src="https://agentmods.dev/badge/skills/vanillagreencom/kendex/youtube-publish/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/vanillagreencom/kendex/youtube-publish"><img src="https://agentmods.dev/badge/skills/vanillagreencom/kendex/youtube-publish.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 | $0.00026 | $0.01052 |
| Opus 5.5 | $0.00010 | $0.00421 |
| Sonnet 5.5 | $0.00005 | $0.00210 |
| Haiku 4.5 | $0.00003 | $0.00105 |
Grade C, and why
youtube-publish 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 17d 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- kendex:project-instructions:start --> How it starts
The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Instructions
Problems with a kendex-owned skill go through kendex report; check ownership in the file first.
YouTube publish
Use Python 3.8 or newer. The executable is scripts/youtube-publish under this installed skill.
Consumer setup
- The consumer supplies
YOUTUBE_CHANNEL_IDandYOUTUBE_APPROVER_EMAILin the command environment. The email identifies the user who can approve a request on Linear. - The consumer supplies
YOUTUBE_ACCESS_TOKENthrough a private credential route. Keep the OAuth client, its secret, refresh token and access token outside the package and version control. An API key does not authorize uploads or comments. - Use the consumer's OAuth client with
youtube.uploadandyoutube.readonlyfor uploads and their channel check. Useyoutube.force-sslfor comments and replies. The consumer's authorization tool owns token acquisition, renewal and revocation. The script checks the token's channel withchannels.listbefore a write. See authorization and channel lookup. - The installed Linear skill supplies live tracker reads. Its configured credentials must be able to read the consumer's tracker item and comment authors.
Draft and approval
- Draft the upload with
draft-upload --issue ID --video PATH --title TEXT --description TEXT --category ID --privacy public|private|unlisted --made-for-kids yes|no. Ask the user to choose the visibility and audience setting. The draft includesstatus.containsSyntheticMedia: true. This discloses realistic altered or synthetic content. See video status. - Draft a top-level comment with
draft-comment --issue ID --video-id ID --text TEXT. For another channel's video, add--target-channel-id IDwith that video's channel. Add--parent-id IDto draft a reply. These commands make no API call. Review the target and exact text with the user. See top-level comments and replies. - Save the JSON draft in the consumer's private working files. Show the user the channel, target, full content, upload visibility and audience setting, and request digest. Each draft prints an
approval_recordobject. The user records their explicit yes by posting that object as the entire comment body on the named Linear item. The comment must come fromYOUTUBE_APPROVER_EMAIL. A caller's assertion or an agent-authored approval is insufficient. - Run
publish --draft PATHonly after that yes. The sending path reads the tracker item and comments live. It checks the author, decision, item and digest before any YouTube request. A changed draft or changed video needs a new approval. Uploads use the documented resumable protocol. - Record the returned resource id and approval comment id on the tracker item. A failed or interrupted write can have an uncertain result. Stop and check the channel with the user before any new attempt. Do not repeat a successful request or reuse its approval for another send.
What ships with it
2 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.
- 17d ago First seen · 49 lines · 26 tokens per session scan C 90220288ee94
youtube-publish is a skill published in the GitHub repository vanillagreencom/kendex (87 stars, last pushed yesterday), licensed MIT. It adds 26 tokens to every session and 1,052 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-12.
Other skills, from other repositories
prose-review
Review prose written for others (e.g., user-facing documentation, prompts for other LLMs, reports, plans, inline comments, docstrings) for local jargon leakage, orphaned references, missing grounding, and audience or genre mismatch. Always run on prose you produce or materially edit, except for routine conversational…
text-search
Search indexed text corpora with qmd. For indexed content, prefer qmd over grep.
rp
Always read this skill when the user mentions "rp" or "repoprompt", or before accessing a repository outside the current RepoPrompt workspace. Covers workspace discovery, binding, root verification, and workspace hygiene.
claude-md-improver
Audit and improve existing CLAUDE.md files across a repository. Use for a deliberate, repo-wide quality review—not to record learnings from the current session.
claude-md-revision
Record durable, project-specific learnings from the current session in CLAUDE.md, .claude.local.md, or AGENTS.md—not general documentation, session summaries, or repo-wide audits.
gather-retrospectives
Gather retrospective feedback from past Pi sessions about a package, tool, or workflow. Use for historical retros and recurring friction, not assessment of only the current session.