← Commit history

Publish 0.8.25

Colby Knox ·c7fff56f6a ·1mo ago ·parent 4716c99
3 files changed +10−5
package.json+1−1
@@ -1,7 +1,7 @@ {   "slug": "adom-wiki-skillpack",   "title": "Adom Wiki Skill Pack",-  "version": "0.8.24",+  "version": "0.8.25",   "type": "skill",   "hero": {     "path": "screenshots/hero.png"
page.json+1−1
@@ -131,7 +131,7 @@       "prompt": "Before I create this wiki page, check whether the thing already has a home."     }   ],-  "version": "0.8.24",+  "version": "0.8.25",   "description": "The Adom Wiki Skill Pack : one main skill plus the focused sub-skills that teach the AI to publish to wiki.adom.inc without breaking pages, name the wiki repo archetypes (Page, Skillpack, Family), and publish rich component pages (chips, molecules, boards). The wiki is six tools in one: git, package manager, releases, discussions, pull requests, and discovery. One package manager install drops the whole set onto your container. Includes reviewed Z-up component artwork, surface-selected MPN marking, reusable thumbnail caches and component-page widgets.",   "dependencies": {     "adom/adom-wiki-hero-image": "^0.2.2",
skills/wiki-discovery/SKILL.md+8−3
@@ -77,13 +77,18 @@ word. A bad value (`has:bogus`) is a 400 `BAD_FILTER` with the allowed set. Filters with no words (`tag:kicad`) list the matching pages, most installed first. Example: `hero studio in:docs owner:adom` finds the section of any Adom page's documents that talks about the studio, and links to the heading.+Search is a locator, not a reader: the file and heading a hit names are where+to START reading, and a broad `in:docs` query can be refused over budget+(`SEARCH_BUDGET`), in which case narrow it with `type:` or `owner:`.  **Search inside one page:** every page has a "Search this page" box in its tab bar (`/` focuses it) backed by `GET /api/pages/<owner>/<slug>/search?q=<words>` (`&family=1` adds the family's children, `&per_file=2` folds sections). It-answers sections with anchors and deep links, so an AI thread can ask "where-in this pack is X documented" without reading the whole pack. Every markdown-document a page ships is indexed by heading at index time; nothing to declare.+answers sections with anchors and deep links, so a thread can find WHICH file+and heading to read first, then read that document. One section per page comes+back and rows are capped, so a miss is a reason to read the pack, not a+conclusion; "not found" never ends a search. Every markdown document a page+ships is indexed by heading at index time; nothing to declare.  ## ⚠️ Current real state (verified 2026-06-21: re-verify, this WILL change)