@ulysses-ai/create-workspace 0.16.0-beta.1 → 0.18.0-beta.0

package/template/.claude/skills/braindump/SKILL.md

@@ -13,7 +13,7 @@ Capture discussion reasoning, exploration results, and design rationale into wor
 
 > **Note:** `/braindump side` has moved to `/aside`. If the user invokes `/braindump side`, redirect them: "The side braindump is now `/aside`. Running it for you." Then invoke the `/aside` skill with their text.
 
-## Session-Aware Behavior
+## Lifecycle-Aware Behavior
 
 When called within an active work session (the active-session pointer at `.claude/.active-session.json` exists inside the current worktree):
 
@@ -26,9 +26,16 @@ When called within an active work session (the active-session pointer at `.claud
   git commit -m "braindump: update {session-name} tracker"
   ```
 
-When called from the workspace root (no active session):
-- Use `--local-only` so the captured file is gitignored (the root only allows local-only writes)
-- Suggest starting a work session if the braindump is about actionable work
+Under the task model — `workspace.sessionModel` is `"task"` in `workspace.json` AND the SessionStart hook injected a `Chat record:` line (`{chat}` is its name):
+
+- Default behavior: write `braindump_{topic}.md` directly into that chat's drawer at `workspace-scratchpad/chats/{chat}/` — the drawer sits outside `workspace-context/`, so `capture-context.mjs` is not involved
+- No commit for drawer writes: the drawer is gitignored and machine-local; `/complete-work` lists it and asks what to promote into `workspace-context/`
+
+When called from the workspace root with no active session — every other case, including a `sessionModel: "session"` workspace (the `Chat record:` line is injected in every chat, so it alone does not select the drawer):
+
+- Use `--local-only` so the captured file is gitignored (the root only allows local-only writes), landing in `team-member/{user}/`
+- If the task model applies but the `Chat record:` line is absent, say the drawer destination is unavailable for that reason
+- Suggest starting work (`/start-work`) if the braindump is about actionable work
 
 The flows below apply when NOT in an active work session, or when the user explicitly asks for a standalone braindump file.
 

package/template/.claude/skills/build-docs-site/SKILL.md

@@ -96,11 +96,11 @@ For the codebase:
 - Read actual implementations, not just existing docs about them
 
 For shared context:
-- Walk `workspace-context/` for handoffs, braindumps, locked team knowledge, release notes
+- Walk `workspace-context/` for handoffs, braindumps, and locked team knowledge
 
 For work-session history:
-- Walk `work-sessions/*/workspace/session.md` for any currently-active session trackers — their bodies may contain decisions not yet consumed into release notes
-- Check git history for previously-completed session trackers that were synthesized into release notes by `/complete-work`
+- Walk `work-sessions/*/workspace/session.md` for any currently-active session trackers — their bodies may contain decisions not yet captured anywhere durable
+- Walk the chat drawers `workspace-scratchpad/chats/*/` for in-progress task-model material — designs, plans, braindumps, research not yet promoted into `workspace-context/` — same rationale: it is thinking the site may need that no merged artifact carries
 
 For existing project documentation (from Phase 1 Q2):
 - Read every file the user pointed at
@@ -312,11 +312,11 @@ Produce a final report:
 - **Files needing user decision.** Any old docs with content that didn't migrate cleanly — coverage check found gaps the user needs to resolve.
 - **URLs that may need redirects.** If the old docs had live URLs, list them. The skill flags but does not set up redirects.
 
-Update the session tracker with the final state. The skill's work ends here. `/complete-work` handles the merge, release notes, and cleanup.
+Update the session tracker with the final state. The skill's work ends here. `/complete-work` handles the merge, the PR, and cleanup.
 
 ## Notes
 
-- Specs and plans live at the project worktree root, not inflight (per workspace-structure rule). They are consumed by `/complete-work` into release notes.
+- Specs and plans live at the project worktree root, not inflight (per workspace-structure rule). They are consumed by `/complete-work` when it builds the PR body, or promoted into `workspace-context/`.
 - Diagrams primitives are deliberately fixed (no project customization) because the class-based fill pattern is load-bearing for theme support.
 - The leak grep is project-accurate because it derives the list from the project's own dependency manifests. Do not maintain a hardcoded master list — it would over-trigger for projects whose docs legitimately discuss their own dependencies.
 - The forbidden-word grep is per-project — each project supplies its own list from Phase 1 Q5. There is no default.

package/template/.claude/skills/build-docs-site/templates/spec.md.tmpl

@@ -116,4 +116,4 @@ Terms that came from brainstorming sessions and need to be translated to descrip
 
 ## Notes
 
-This spec is consumed by `/complete-work` into release notes when the session finishes. Update it as decisions change during writing.
+This spec feeds the PR body when `/complete-work` finishes the session (or is promoted into `workspace-context/`). Update it as decisions change during writing.