@leeovery/claude-technical-workflows 2.1.30 → 2.1.32

package/skills/start-specification/references/handoffs/unify.md

@@ -4,6 +4,21 @@
 
 ---
 
+Before invoking the processing skill, save a session bookmark.
+
+> *Output the next fenced block as a code block:*
+
+```
+Saving session state so Claude can pick up where it left off if the conversation is compacted.
+```
+
+```bash
+.claude/hooks/workflows/write-session-state.sh \
+  "unified" \
+  "skills/technical-specification/SKILL.md" \
+  "docs/workflow/specification/unified/specification.md"
+```
+
 This skill's purpose is now fulfilled. Invoke the [technical-specification](../../../technical-specification/SKILL.md) skill for your next instructions. Do not act on the gathered information until the skill is loaded — it contains the instructions for how to proceed.
 
 ```

package/skills/status/SKILL.md

@@ -3,6 +3,12 @@ name: status
 description: "Show workflow status - what exists, where you are, and what to do next."
 disable-model-invocation: true
 allowed-tools: Bash(.claude/skills/status/scripts/discovery.sh)
+hooks:
+  PreToolUse:
+    - hooks:
+        - type: command
+          command: "$CLAUDE_PROJECT_DIR/.claude/hooks/workflows/system-check.sh"
+          once: true
 ---
 
 Show the current state of the workflow for this project.
@@ -13,7 +19,7 @@ Show the current state of the workflow for this project.
 
 Invoke the `/migrate` skill and assess its output.
 
-**If files were updated**: STOP and wait for the user to review the changes (e.g., via `git diff`) and confirm before proceeding to Step 1.
+**If files were updated**: STOP and wait for the user to review the changes (e.g., via `git diff`) and confirm before proceeding to Step 1. Do not continue automatically.
 
 **If no updates needed**: Proceed to Step 1.
 

package/skills/technical-planning/SKILL.md

@@ -62,7 +62,7 @@ Context refresh (compaction) summarizes the conversation, losing procedural deta
 2. **Read all tracking and state files** for the current topic — plan index files, review tracking files, implementation tracking files, or any working documents this skill creates. These are your source of truth for progress.
 3. **Check git state.** Run `git status` and `git log --oneline -10` to see recent commits. Commit messages follow a conventional pattern that reveals what was completed.
 4. **Announce your position** to the user before continuing: what step you believe you're at, what's been completed, and what comes next. Wait for confirmation.
-5. **Check `author_gate_mode` and `finding_gate_mode`** in the Plan Index File frontmatter — if `auto`, the user previously opted in during this session. Preserve these values.
+5. **Check `task_list_gate_mode`, `author_gate_mode`, and `finding_gate_mode`** in the Plan Index File frontmatter — if `auto`, the user previously opted in during this session. Preserve these values.
 
 Do not guess at progress or continue from memory. The files on disk and git history are authoritative — your recollection is not.
 
@@ -118,7 +118,7 @@ Found existing plan for **{topic}** (previously reached phase {N}, task {M}).
 
 If the specification changed, update `spec_commit` in the Plan Index File frontmatter to the current commit hash.
 
-Reset `author_gate_mode` and `finding_gate_mode` to `gated` in the Plan Index File frontmatter (fresh invocation = fresh gates).
+Reset `task_list_gate_mode`, `author_gate_mode`, and `finding_gate_mode` to `gated` in the Plan Index File frontmatter (fresh invocation = fresh gates).
 
 → Proceed to **Step 1**.
 

package/skills/technical-planning/references/author-tasks.md

@@ -37,6 +37,8 @@ Check `author_gate_mode` in the Plan Index File frontmatter.
 
 #### If `author_gate_mode: auto`
 
+**Auto mode removes the approval pause — not the sequential process.** Each task is still invoked, authored, and logged one at a time, in order. Do not batch, skip ahead, or create multiple tasks concurrently.
+
 > *Output the next fenced block as a code block:*
 
 ```
@@ -55,7 +57,7 @@ Task {M} of {total}: {Task Name} — authored. Logging to plan.
 · · · · · · · · · · · ·
 **To proceed:**
 - **`y`/`yes`** — Approved. I'll log it to the plan.
-- **`a`/`auto`** — Approve this and all remaining tasks automatically
+- **`a`/`auto`** — Approve this and all remaining task authoring gates automatically
 - **Or tell me what to change.**
 - **Or navigate** — a different phase or task, or the leading edge.
 · · · · · · · · · · · ·

package/skills/technical-planning/references/define-tasks.md

@@ -43,7 +43,31 @@ planning:
 
 Commit: `planning({topic}): draft Phase {N} task list`
 
-Present the task overview to the user as rendered markdown (not in a code block). Then, separately, present the choices:
+Present the task overview to the user:
+
+> *Output the next fenced block as markdown (not a code block):*
+
+```
+{task overview from planning-task-designer agent}
+```
+
+Then check the gate mode.
+
+### Check Gate Mode
+
+Check `task_list_gate_mode` in the Plan Index File frontmatter.
+
+#### If `task_list_gate_mode: auto`
+
+> *Output the next fenced block as a code block:*
+
+```
+Phase {N}: {Phase Name} — task list approved. Proceeding to authoring.
+```
+
+→ Skip to **If approved** below.
+
+#### If `task_list_gate_mode: gated`
 
 **STOP.** Ask:
 
@@ -53,6 +77,7 @@ Present the task overview to the user as rendered markdown (not in a code block)
 · · · · · · · · · · · ·
 **To proceed:**
 - **`y`/`yes`** — Approved.
+- **`a`/`auto`** — Approve this and all remaining task list gates automatically
 - **Or tell me what to change** — reorder, split, merge, add, edit, or remove tasks.
 - **Or navigate** — a different phase or task, or the leading edge.
 · · · · · · · · · · · ·
@@ -66,12 +91,19 @@ Re-invoke `planning-task-designer` with all original inputs PLUS:
 
 Update the Plan Index File with the revised task table, re-present, and ask again. Repeat until approved.
 
-#### If approved
+#### If `auto`
+
+Note that `task_list_gate_mode` should be updated to `auto` during the commit step below.
+
+→ Proceed to **If approved** below.
+
+#### If approved (`y`/`yes` or `auto`)
 
 **If the task list is new or was amended:**
 
 1. Advance the `planning:` block to the first task in this phase
-2. Commit: `planning({topic}): approve Phase {N} task list`
+2. If user chose `auto` at this gate: update `task_list_gate_mode: auto` in the Plan Index File frontmatter
+3. Commit: `planning({topic}): approve Phase {N} task list`
 
 **If the task list was already approved and unchanged:** No updates needed.
 

package/skills/technical-planning/references/output-formats/tick/about.md

@@ -54,6 +54,11 @@ Add to `.gitignore`:
 
 The 3-level hierarchy (topic → phase → task) uses Tick's parent/child system. Parent tasks are implicitly blocked by their children — a parent is not "ready" until all children are complete. Explicit dependencies (`tick dep add`) handle cross-phase and cross-topic blocking.
 
+## Usage Notes
+
+- **Help**: `tick help --all` displays all available commands and flags in a single view.
+- **Output format**: The `--toon` flag is **not needed** — TOON is the default for non-interactive shells, which is what Claude Code uses. Omit it from all commands.
+
 ## Output Location
 
 Tasks are stored in a `.tick/` directory at the project root:

package/skills/technical-planning/references/output-formats/tick/authoring.md

@@ -1,5 +1,16 @@
 # Tick: Authoring
 
+## Sandbox Mode and Large Descriptions
+
+Bash heredocs (`$(cat <<'EOF'...EOF)`) create temp files that sandbox mode blocks, resulting in empty descriptions being set silently. Do **not** use `dangerouslyDisableSandbox` — it forces user approval on every call.
+
+Instead, use the **Write tool + cat pattern**:
+
+1. Use the **Write tool** to save the description to `$TMPDIR/tick-desc.txt` — this bypasses sandbox because it uses Claude Code's file writing, not bash
+2. Run the tick command with `--description "$(cat $TMPDIR/tick-desc.txt)"` in normal sandbox mode — `cat` just reads, no temp files needed
+
+This works for both `tick create --description` and `tick update --description`.
+
 ## Task Storage
 
 Tasks are created using the `tick create` command. Before creating individual tasks, establish the topic and phase parent tasks.

package/skills/technical-planning/references/output-formats/tick/updating.md

@@ -16,6 +16,8 @@ Tick uses dedicated commands for each status transition:
 
 ## Updating Task Content
 
+**Sandbox mode**: When updating large descriptions, use the Write tool + cat pattern to avoid sandbox temp file issues. See [authoring.md](authoring.md) for details.
+
 To update a task's properties:
 
 - **Title**: `tick update <task-id> --title "New title"`

package/skills/technical-planning/references/plan-construction.md

@@ -79,6 +79,20 @@ After Step A returns with an approved task table, continue to **Author Tasks for
 {task list from the phase's task table}
 ```
 
+Check `task_list_gate_mode` in the Plan Index File frontmatter.
+
+#### If `task_list_gate_mode: auto` (existing task table)
+
+> *Output the next fenced block as a code block:*
+
+```
+Phase {N}: {Phase Name} — task list confirmed. Proceeding to authoring.
+```
+
+→ Continue to **Author Tasks for the Phase** below.
+
+#### If `task_list_gate_mode: gated` (existing task table)
+
 > *Output the next fenced block as markdown (not a code block):*
 
 ```
@@ -106,16 +120,11 @@ After Step A returns with an approved task table, continue to **Author Tasks for
 
 Work through each task in the phase's task table, in order.
 
-### Parallel authoring (optional optimization)
-
-After the first `pending` task in a phase is approved, you may invoke multiple Step B agents concurrently for tasks you judge to be independent — where the authored detail of one would not inform the other. This is an invocation optimization only; the approval flow is unchanged:
+### Sequential authoring (mandatory)
 
-- Present tasks one at a time, in order
-- Each task still requires explicit user approval before logging
-- If user feedback on a presented task changes context that could affect any already-authored task waiting to be presented, discard those results and re-invoke Step B
-- When uncertain about independence, default to sequential — it is always safe
+Tasks **must** be authored one at a time, in order. Each task's authored detail builds context for the next — later tasks benefit from decisions, patterns, and structure established by earlier ones. This applies regardless of gate mode. Auto-gate removes the approval pause, not the sequential process.
 
-Never parallelize the first `pending` task in a phase. Never parallelize across phases.
+**Never** invoke multiple Step B agents concurrently. **Never** batch or skip ahead.
 
 #### If the task status is `authored`
 

package/skills/technical-planning/references/plan-index-schema.md

@@ -23,6 +23,7 @@ spec_commit: {commit-hash}
 created: YYYY-MM-DD
 updated: YYYY-MM-DD
 external_dependencies: []
+task_list_gate_mode: gated
 author_gate_mode: gated
 finding_gate_mode: gated
 planning:
@@ -43,6 +44,7 @@ planning:
 | `created` | Plan creation — today's date |
 | `updated` | Plan creation — today's date; update on each commit |
 | `external_dependencies` | Dependency resolution (Step 6) |
+| `task_list_gate_mode` | Plan creation → `gated`; user opts in → `auto` |
 | `author_gate_mode` | Plan creation → `gated`; user opts in → `auto` |
 | `finding_gate_mode` | Plan creation → `gated`; user opts in → `auto` |
 | `planning.phase` | Tracks current phase position |
@@ -88,14 +90,15 @@ approved_at: {YYYY-MM-DD}
 
 ```markdown
 #### Tasks
-| ID | Name | Status | Ext ID |
-|----|------|--------|--------|
-| {topic}-{phase}-{seq} | {Task Name} | {status} | |
+| ID | Name | Edge Cases | Status | Ext ID |
+|----|------|------------|--------|--------|
+| {topic}-{phase}-{seq} | {Task Name} | {comma-separated list, or "none"} | {status} | |
 ```
 
 | Field | Set when |
 |-------|----------|
 | `ID` | Task design — format: `{topic}-{phase}-{seq}` |
 | `Name` | Task design — descriptive task name |
+| `Edge Cases` | Task design — curated list scoping which edge cases this task handles |
 | `Status` | Task design → `pending`; authoring → `authored` |
 | `Ext ID` | Task authored — external identifier for the task |

package/skills/view-plan/SKILL.md

@@ -2,6 +2,12 @@
 name: view-plan
 description: "View a plan's tasks and progress, regardless of output format."
 disable-model-invocation: true
+hooks:
+  PreToolUse:
+    - hooks:
+        - type: command
+          command: "$CLAUDE_PROJECT_DIR/.claude/hooks/workflows/system-check.sh"
+          once: true
 ---
 
 Display a readable summary of a plan's phases, tasks, and status.