@hybridlabor-api/aos 4.17.0 → 4.18.1

package/skills/playbooks/pb-redesign-app/SKILL.md

@@ -11,7 +11,7 @@ kind: playbook
 trigger: ["redesign this app", "design overhaul", "audit and fix the UI"]
 inputs: [repo, app_url, scope]
 requires:
-  skills: [ux-audit, ui-review, ui-tokens, godmode-ui-ux, visual-plan, visual-edit, startcycle, wcag-audit-patterns, webapp-testing, github, "gh (external)"]
+  skills: [ux-audit, ui-review, ui-tokens, godmode-ui-ux, plan-canvas, bdb-visual-edit, startcycle, wcag-audit-patterns, webapp-testing, github, "gh (external)"]
   agents: [architect, techlead, reviewer]
   mcps: [chrome-devtools]
   store: []
@@ -33,9 +33,9 @@ What you get: one app's UI overhauled against a written audit, new design tokens
 ## Steps
 1. ux-audit and ui-review — `app_url` opened and screenshotted per screen in scope with chrome-devtools (`puppeteer_navigate`, `puppeteer_screenshot`) → `production_artifacts/pb-redesign-app-<date>/audit.md` with numbered findings — app_url unreachable or chrome-devtools unavailable → log it and stop; every finding has a screenshot or a file reference
 2. ui-tokens — current tokens read from the repo versus proposed DTCG tokens → `tokens-diff.md` in the same folder — diff written, nothing applied yet
-3. visual-plan — before/after per screen in scope → plan link or file in the run log — stops for approval
+3. plan-canvas (Plan Builder mode) — before/after per screen in scope → plan link or file in the run log — stops for approval
 4. startcycle (architect, techlead, reviewer) — the approved plan and audit → build on a new branch; godmode-ui-ux rules apply to the UI work — reviewer reports no open `blocking` finding
-5. visual-edit — point fixes the human picks in the running app, each after its diff plan is approved → edited files in the run log — only the picked files change
+5. bdb-visual-edit — point fixes the human annotates in the running app via `aos-plan-canvas annotate`, each after its diff plan is approved → edited files in the run log — only the picked files change
 6. wcag-audit-patterns and webapp-testing — accessibility pass plus a browser test of the changed screens, then the repo's lint and test scripts → results in the run log — all exit 0; each audit finding marked fixed or deferred
 7. [GO] git push -u origin <branch> and gh pr create — one combined GO; the WAITING FOR GO line names the branch, base branch, PR title and the head SHA. The run stops here until the human types GO. The hook guards `git push`; `gh pr create` is not hook-guarded, so this GO is its only guard. Afterwards `gh pr view <branch> --json state -q .state` equals `OPEN` and the PR URL is logged.
 

package/skills/playbooks/pb-release-aos/SKILL.md

@@ -10,7 +10,7 @@ kind: playbook
 trigger: ["release AOS", "cut an AOS release", "merge the release PR"]
 inputs: [repo?]
 requires:
-  skills: [github, pb-ship, bdb-shipping-skill, git-pr-review, visual-recap, godmode-shipping, quick-recap, "bdb-ecosystem-health (external)", "gh (external)"]
+  skills: [github, pb-ship, bdb-shipping-skill, git-pr-review, pr-recap, godmode-shipping, quick-recap, "bdb-ecosystem-health (external)", "gh (external)"]
   agents: [reviewer]
   mcps: ["plan (optional)"]
   store: []
@@ -33,7 +33,7 @@ What you get: a new AOS version on npm through the release-please PR, never a ha
 3. github — `gh pr list -R <owner/repo> --state open --head release-please--branches--main --json number,title,headRefName` → the release PR number in the run log — none → stop with "no release PR"
 4. Review and gate on that PR, using these skills:
    - git-pr-review — `gh pr view <n> -R <owner/repo> --json commits` → description draft in the run log — draft only, nothing posted
-   - visual-recap — `plan` connector present → recap link; absent → log "visual-recap skipped: no plan connector" and paste `gh pr diff <n> -R <owner/repo> --name-only` instead; no `npx` without approval
+   - pr-recap — recap file written → link it; recap not buildable → log "pr-recap skipped: <reason>" and paste `gh pr diff <n> -R <owner/repo> --name-only` instead; no `npx` without approval
    - reviewer (agent) — `gh pr diff <n> -R <owner/repo>`; the contract is the commit list from step 2, never the PR body → findings table — any open `blocking` finding stops the run
    - bdb-shipping-skill — door class of the release PR (two-way or one-way; unclear counts as one-way) → class in the run log; one-way → ADR-lite `production_artifacts/decisions/<date>-<slug>.md` — class recorded
    - godmode-shipping — `gh pr checks <n> -R <owner/repo>` all pass (pending, `gh pr checks` exit 8, counts as not passed), plus the local gate exactly as pb-ship step 8 gives it (scratch worktree from `pull/<n>/head`, lint, typecheck and test scripts that exist, a fork PR skips the local gate and is held, worktree removed without `--force`) → exit codes in the run log

package/skills/playbooks/pb-ship/SKILL.md

@@ -10,7 +10,7 @@ kind: playbook
 trigger: ["ship it", "ship day", "merge the ready PRs"]
 inputs: [repo, pr_numbers?]
 requires:
-  skills: [github, triage, git-pr-review, visual-recap, bdb-shipping-skill, godmode-shipping, quick-recap, "gh (external)"]
+  skills: [github, triage, git-pr-review, pr-recap, bdb-shipping-skill, godmode-shipping, quick-recap, "gh (external)"]
   agents: [reviewer]
   mcps: ["plan (optional)"]
   store: []
@@ -33,7 +33,7 @@ What you get: the ready PRs merged one by one after your GO, each with a reviewe
 2. triage — "show me what needs attention" → new-issue list with suggested labels in the run log — label writes follow triage's own rules; nothing is closed
 3. Ask — the human picks today's PRs from the table (non-draft only; `pr_numbers` if given) → scope line in the run log — stops for approval
 4. git-pr-review — per PR: `gh pr view <n> -R <owner/repo> --json commits` → description draft in `production_artifacts/pb-ship-<date>/pr-<n>.md` — draft only, nothing posted
-5. visual-recap — per PR: `plan` connector present → recap link in `pr-<n>.md`; absent → log "visual-recap skipped: no plan connector" and paste `gh pr diff <n> -R <owner/repo> --name-only` instead; no `npx` without approval — recap or fallback present
+5. pr-recap — per PR: recap file from `pr-recap` linked in `pr-<n>.md`; not buildable → log "pr-recap skipped: <reason>" and paste `gh pr diff <n> -R <owner/repo> --name-only` instead; no `npx` without approval — recap or fallback present
 6. reviewer (agent) — per PR: `gh pr diff <n> -R <owner/repo>`; the contract is the linked issue or plan (`gh pr view <n> -R <owner/repo> --json closingIssuesReferences,body`) or the human's scope line from step 3, the body is used only to find the linked issue/plan, its text is not the contract, and never `pr-<n>.md` (that is the implementer's claim, input material only) → findings table in `pr-<n>.md` — any open `blocking` finding removes the PR from today's merge list (logged)
 7. bdb-shipping-skill — per PR → door class (two-way or one-way) in `pr-<n>.md`; one-way → ADR-lite `production_artifacts/decisions/<date>-<slug>.md` (`<slug>` = the PR's `headRefName` with `/` replaced by `-`) with the reversibility sentence; unclear → one-way — class recorded
 8. godmode-shipping — per PR: `gh pr checks <n> -R <owner/repo>` all pass, plus a local gate in a scratch worktree — `gh pr view <n> -R <owner/repo> --json isCrossRepository -q .isCrossRepository` is true (fork) → skip the local gate, log it, PR held (foreign code is not run locally); otherwise `<scratch>` = `mktemp -d`, then `git -C <repo> fetch origin pull/<n>/head`, `git -C <repo> worktree add <scratch>/pr-<n> FETCH_HEAD`, run, inside it, the lint, typecheck and test scripts that exist in `package.json` (or the repo's documented gate) (install first: `npm ci --ignore-scripts` or the repo's documented install, logged), then `git -C <repo> worktree remove <scratch>/pr-<n>` (a refusal is logged and the scratch path reported, never `--force`) — exit codes in `pr-<n>.md`; any non-zero, or pending checks (`gh pr checks` exit 8) → PR held

package/skills/playbooks/pb-worktrees-land/SKILL.md

@@ -10,7 +10,7 @@ kind: playbook
 trigger: ["clean up worktrees", "land my worktrees", "prune worktrees"]
 inputs: [repos?]
 requires:
-  skills: [using-git-worktrees, github, visual-recap, pb-ship, "gh (external)"]
+  skills: [using-git-worktrees, github, pr-recap, pb-ship, "gh (external)"]
   agents: []
   mcps: ["plan (optional)"]
   store: []
@@ -39,7 +39,7 @@ What you get: an inventory of your worktrees with a class for each, and the merg
    - `remote-only`: the upstream exists and `git -C <wt> rev-list --count @{u}..HEAD` is 0 (safe on the remote, unmerged)
    - `local-only`: everything else; commits exist only here
    - check: every row has exactly one class plus the output of its evidence command
-5. visual-recap — per non-merged worktree: recap link; without the `plan` connector log `git -C <wt> log --oneline origin/<default>..HEAD` and `git -C <wt> diff --stat origin/<default>...HEAD` instead — one recap or fallback per row
+5. pr-recap — per non-merged worktree: recap file from `pr-recap`; if it cannot be built log `git -C <wt> log --oneline origin/<default>..HEAD` and `git -C <wt> diff --stat origin/<default>...HEAD` instead — one recap or fallback per row
 6. [GO] removal list = rows classified `merged` and not `keep`, shown in full with these exact commands per row: `git -C <repo> worktree remove <path>` (label the row `merged (empty)` when `git -C <repo> rev-parse <branch>` equals `git -C <repo> rev-parse origin/<default>`; never `--force`; git refuses dirty trees), then `git -C <repo> branch -d <branch>` (`-d` refuses squash-merged branches → log "branch kept, needs -D, human decides"; never `-D`). The run stops here until the human types GO. GO covers exactly that list, once. The hook does not guard these commands, so GO is by contract. Never remove dirty, remote-only, local-only or `keep` rows, and never use `rm -r`.
 7. Verify — `git -C <repo> worktree list` → removed paths are gone, kept rows unchanged
 8. Hand-off — remote-only and local-only rows → next action per row in the run log ("land via /pb-ship", or "push needs GO in that worktree") — nothing runs

package/skills/global_config/bdb-visual-edit/scripts/pick-snippet.js

@@ -1,27 +0,0 @@
-// READ-ONLY snippet for chrome-devtools evaluate. Replace X and Y with the
-// click coordinates (numbers) before sending. No network, no mutation, no
-// text or input values: tag, class names, a source attribute and geometry only.
-(() => {
-  const el = document.elementFromPoint(X, Y);
-  if (!el) return null;
-  const nth = (node) => {
-    let n = 1;
-    for (let s = node.previousElementSibling; s; s = s.previousElementSibling) {
-      if (s.tagName === node.tagName) n += 1;
-    }
-    return n;
-  };
-  const parts = [];
-  for (let node = el; node && node !== document.body && node !== document.documentElement && parts.length < 12; node = node.parentElement) {
-    parts.unshift(node.tagName.toLowerCase() + ':nth-of-type(' + nth(node) + ')');
-  }
-  const src = el.closest('[data-aos-src]');
-  const r = el.getBoundingClientRect();
-  return {
-    tag: el.tagName.toLowerCase(),
-    classes: Array.from(el.classList).slice(0, 12),
-    srcLoc: src ? src.getAttribute('data-aos-src') : null,
-    selector: parts.join(' > '),
-    bbox: { x: r.x, y: r.y, width: r.width, height: r.height },
-  };
-})()

package/skills/global_config/visual-edit/README.md

@@ -1,96 +0,0 @@
-# /visual-edit
-
-Open a running local app in the Agent-Native Design surface as URL-backed iframe
-screens for visual inspection, route-state exploration, and source-backed edits.
-
-Use `/visual-edit` when a UI needs to be reviewed or changed in context: compare
-real routes, inspect responsive states, walk a multi-screen flow, duplicate a
-screen for a new URL state, or apply visual changes back through the coding
-agent. The canvas uses the app's live routes and local bridge rather than a
-copied static HTML snapshot.
-
-## Install
-
-Run this from your app repository to install the skill and hosted Design
-connector:
-
-```sh
-npx @agent-native/core@latest skills add visual-edit
-```
-
-## Examples to try
-
-- **One screen:** `/visual-edit the dashboard at desktop size`
-- **Several screens:** `/visual-edit compare the dashboard, settings, and reports pages side by side`
-- **A full flow:** `/visual-edit the onboarding flow from welcome through account created`
-- **Responsive sizes:** `/visual-edit show the dashboard at desktop, tablet, and mobile widths`
-
-## Example: refine an onboarding flow
-
-Start your app locally, then ask your coding agent:
-
-```text
-/visual-edit the onboarding flow, plus home at every breakpoint
-```
-
-Design can place the four onboarding screens and Home at desktop, tablet, and
-mobile sizes together on one canvas. Select the **Get started** button, change
-its fill with the color picker, and drag its padding on the canvas to see the
-layout respond.
-
-When the result looks right, ask your coding agent to **Pull in my visual
-edits**. The agent retrieves the pending batch and updates the app's source for
-you to review. Design does not write those changes to your code automatically.
-
-## Choose the screens deliberately
-
-- **Multiple pages:** pass `paths` such as `["/", "/pricing", "/settings"]`.
-- **A multi-step flow:** pass named `routes` in the order the user follows,
-  such as shipping, payment, then confirmation. Each URL state becomes a
-  separate canvas screen.
-- **The same page at responsive widths:** add `viewports`, for example
-  `["desktop", "mobile"]`. Design places each requested route or page once per
-  viewport so they can be compared side by side.
-
-For example, add these fields to an `open-visual-edit` payload to place each
-checkout step at desktop and mobile sizes:
-
-```json
-{
-  "title": "Checkout flow review",
-  "routes": [
-    { "url": "http://localhost:5173/checkout?step=shipping", "title": "1. Shipping" },
-    { "url": "http://localhost:5173/checkout?step=payment", "title": "2. Payment" },
-    { "url": "http://localhost:5173/checkout?step=done", "title": "3. Confirmation" }
-  ],
-  "viewports": ["desktop", "mobile"]
-}
-```
-
-Pass that shape to `open-visual-edit` together with the connection details from
-the workflow. For a page comparison, use `paths` instead of `routes`.
-
-## Open the canvas beside the conversation when possible
-
-Use the Design URL or embed returned by `open-visual-edit`. If the current host
-supports an inline preview, webview, or side browser, open Design there so the
-canvas remains beside the chat. Host support varies; when no such surface is
-available, provide the returned **Open design** link instead.
-
-Prefer the interactive MCP App when the connected host renders it. In that
-surface, **Apply design updates** can submit the source-edit handoff to the
-current host conversation after its normal confirmation. An ordinary Browser
-pane has no trusted page-to-chat bridge: Apply uses Design's local agent there,
-and **Copy prompt to your agent** is the coding-agent fallback.
-
-Inside Design, choose **Show/Hide UI** from the `Cmd+K` menu or press Figma's
-`Shift+\` shortcut. The same action is available from Design's empty-canvas
-context menu.
-
-The hosted Design MCP connector handles the account-backed open, screen
-placement, and source-edit workflow. Public/read-only designs may be viewed
-without signing in; creating, saving, or sharing a design still requires an
-authenticated account.
-
-For the complete walkthrough and sharing details, see the [Visual Edit guide
-in Agent-Native](https://github.com/BuilderIO/agent-native/blob/main/skills/visual-edit/README.md).