One .zip per skill. Inside: SKILL.md with the intent and the frontmatter, and the code that runs in the page.
Settings → Personalize → Skills. This is Cowork's own skill library, shared across every chat.
Click Add and choose the file you just downloaded. Cowork installs it once; no CLAUDE.md, no per-project setup.
The skill stays available in every session. Cowork runs it in your own Chrome, riding the session you are already signed into.
Search YouTube videos by keyword, returning title, channel, views and length.
{"query":"string, REQUIRED (search text)"}{"count":"int","query":"string","results":"array of { videoId, title, channel, url, views, published, length }"}Searches videos by keyword. Each result's videoId is what youtube-get-video takes, the first step of the chain. Resolves in a single bridge call: the extractor fetches what it needs itself, so the params actually drive the request and no page is opened.
The extractor lives as text in the site's localStorage (key youtube-search-videos) and is rebuilt on each call with synchronous eval. It is stored once and then persists across navigation and browser restarts, so on every run except the first it is already there.
Default action: just call (step 1). The call itself reports NEEDS_STORE when the extractor is missing or out of date, only then do you store it (step 2) and call again.
Versioning. The stored extractor is gated by VER (must match the version in the frontmatter above). Both snippets carry VER = "1", keep them in sync with the frontmatter. When evaluate.js changes, bump version and VER in both snippets: the next call sees the mismatch, returns NEEDS_STORE, and re-stores automatically. Storage stays at two keys (youtube-search-videos and youtube-search-videos:ver), overwritten each store, versions never accumulate.
await (() => {
const KEY = "youtube-search-videos", VER = "1";
const s = localStorage.getItem(KEY);
if (!s || localStorage.getItem(KEY + ":ver") !== VER) return 'NEEDS_STORE';
const fn = eval('(' + s + ')');
return fn(document, { query: "claude code" });
})()
Returns the matching videos (~18 per call). If it returns the string 'NEEDS_STORE', the extractor is missing or stale in this browser → do step 2 once, then run this exact call again.
Run this snippet verbatim. Never add an await before the eval, only the outer await is allowed.
Any youtube.com tab works, the extractor never reads the open page; the tab only supplies the origin and session.
NEEDS_STORE; persists until localStorage is cleared)From any youtube.com tab, evaluate the following with the bridge, replacing <FN> with the contents of evaluate.js (the ( async (root, params) => { ... } ) function). Keep VER equal to the call snippet's:
(() => {
const KEY = "youtube-search-videos", VER = "1";
const fn = (<FN>);
localStorage.setItem(KEY, String(fn));
localStorage.setItem(KEY + ":ver", VER);
return 'stored';
})()
Returns "stored". The extractor passes through the model only here, once. Then go back to step 1.
views, published and length are display strings in the session's language, an es session answers "570,944 vistas" / "hace 3 meses" (verified). The extractor asks for hl=en, but the account's own preference overrides it, so do not rely on English. Treat those three as opaque text; videoId is the field to build on, and youtube-get-video returns the exact numeric views for any id.
NEEDS_STORE → extractor missing or version-mismatched in this browser; run step 2 once, then retry step 1.missing_params: query is required → pass a search text.blocked → a consent wall; open youtube.com in the tab, accept once, retry.tool-broken: no ytInitialData → YouTube changed its bootstrap; recapture.request_failed / fetch fail → no youtube.com tab open, or the call ran cross-origin.{ "type": "json", "jsonPath": "count" }