sequant 1.2.5 → 1.3.1

package/templates/skills/fullsolve/SKILL.md

@@ -330,6 +330,23 @@ gh pr merge <N> --squash --delete-branch
 
 **Why this order matters:** The cleanup script checks if the PR is merged before proceeding. The `--delete-branch` flag will fail to delete the local branch (worktree conflict) but successfully deletes the remote branch. The cleanup script then handles the local branch removal.
 
+### 5.4 Post-Merge Verification
+
+**Recommended:** After merge, verify the build and CLI still work:
+
+```bash
+# Pull latest main
+git pull origin main
+
+# Rebuild and verify
+npm run build
+
+# Smoke test - verify CLI runs without errors
+npx sequant doctor
+```
+
+If any command fails, fix immediately on main before continuing. This catches issues like ESM compatibility bugs that unit tests may miss.
+
 ## Iteration Tracking
 
 Track iterations to prevent infinite loops:

package/templates/skills/spec/references/recommended-workflow.md

@@ -0,0 +1,92 @@
+# Recommended Workflow Format
+
+This document shows the expected output format for the `## Recommended Workflow` section in `/spec` output. The `parseRecommendedWorkflow()` function parses this format to determine which phases to execute.
+
+## Format
+
+```markdown
+## Recommended Workflow
+
+**Phases:** spec → exec → qa
+**Quality Loop:** disabled
+**Reasoning:** Brief explanation of why this workflow was chosen.
+```
+
+## Examples
+
+### Simple Bug Fix
+
+```markdown
+## Recommended Workflow
+
+**Phases:** exec → qa
+**Quality Loop:** disabled
+**Reasoning:** Straightforward bug fix with clear root cause. No planning needed.
+```
+
+### Standard Feature
+
+```markdown
+## Recommended Workflow
+
+**Phases:** spec → exec → qa
+**Quality Loop:** disabled
+**Reasoning:** New feature with defined scope. Standard workflow applies.
+```
+
+### UI Feature
+
+```markdown
+## Recommended Workflow
+
+**Phases:** spec → exec → test → qa
+**Quality Loop:** enabled
+**Reasoning:** UI feature requires browser testing and may need iteration.
+```
+
+### Security-Sensitive Feature
+
+```markdown
+## Recommended Workflow
+
+**Phases:** spec → security-review → exec → qa
+**Quality Loop:** disabled
+**Reasoning:** Auth-related changes require security analysis before implementation.
+```
+
+### Complex Refactor
+
+```markdown
+## Recommended Workflow
+
+**Phases:** spec → exec → test → qa
+**Quality Loop:** enabled
+**Reasoning:** Complex refactor with UI components requires browser testing and iteration.
+```
+
+## Phase Separators
+
+The parser supports multiple separator formats:
+
+- Arrow: `spec → exec → qa`
+- ASCII arrow: `spec -> exec -> qa`
+- Comma: `spec, exec, qa`
+
+## Quality Loop Values
+
+The parser accepts these values for the Quality Loop setting:
+
+- Enabled: `enabled`, `true`, `yes`
+- Disabled: `disabled`, `false`, `no`
+
+## Available Phases
+
+| Phase | Description |
+|-------|-------------|
+| `spec` | Plan review and verification criteria generation |
+| `security-review` | Deep security analysis for sensitive features |
+| `testgen` | Generate test stubs from specification |
+| `exec` | Implementation of the feature |
+| `test` | Browser-based testing for UI features |
+| `qa` | Code review against acceptance criteria |
+| `loop` | Quality iteration loop for fixing issues |

package/templates/skills/test/SKILL.md

@@ -8,13 +8,13 @@ metadata:
 allowed-tools:
   - Read
   - Bash
-  - mcp__chrome-devtools__*
+  - mcp__chrome-devtools__*  # Optional: falls back to manual checklist if unavailable
   - Glob
   - Grep
   - TodoWrite
   - Bash(gh issue view:*)
   - Bash(gh issue comment:*)
-  - Bash(npm run dev:*)
+  - Bash({{PM_RUN}} dev:*)
   - Bash(lsof:*)
   - Bash(npx tsx:*)
 ---
@@ -101,18 +101,29 @@ Check for test data requirements:
 
 ### 1.4 Dev Server Check
 
-Check if dev server is running:
+**Extract port from DEV_URL configuration:**
+The dev server URL is configured in `.claude/.sequant/config.json` under `tokens.DEV_URL`. Extract the port for the `lsof` check:
+
 ```bash
-lsof -ti:3000
+# Get DEV_URL from config (default: {{DEV_URL}})
+# Extract port: http://localhost:PORT -> PORT
+DEV_PORT=$(echo "{{DEV_URL}}" | sed -E 's/.*:([0-9]+).*/\1/')
+
+# Check if dev server is running on configured port
+lsof -ti:$DEV_PORT
 ```
 
-If not running, start it:
+If not running, start it using the project's package manager:
 ```bash
-npm run dev
+{{PM_RUN}} dev
 ```
 
 Wait for server ready before proceeding.
 
+**Note:** If `{{DEV_URL}}` or `{{PM_RUN}}` are not replaced with actual values, the defaults are:
+- DEV_URL: `http://localhost:3000` (Next.js), `http://localhost:4321` (Astro), `http://localhost:5173` (Vite-based)
+- PM_RUN: `npm run` (or `bun run`, `yarn`, `pnpm run` based on lockfile)
+
 ## Decision Point: Feature Implemented or Not?
 
 At this point, you've checked if the feature exists (section 1.2). Based on that result:
@@ -240,6 +251,62 @@ When a bug is discovered during testing:
    - Mark as PASS/FAIL based on fix
    - Continue with remaining tests
 
+### 2.4 MCP Availability Check (Graceful Fallback)
+
+**Before starting browser automation**, check if Chrome DevTools MCP is available:
+
+```
+Check if mcp__chrome-devtools__* tools are available in your current session.
+```
+
+**If MCP IS available:**
+- Proceed with automated browser testing (Phase 2.1-2.3)
+
+**If MCP is NOT available:**
+- Skip browser automation steps
+- Generate a **Manual Testing Checklist** instead
+
+**Manual Testing Checklist (No MCP Fallback):**
+
+When browser automation is unavailable, generate a structured manual testing guide:
+
+```markdown
+## Manual Testing Checklist for Issue #<N>
+
+**Pre-requisites:**
+- [ ] Dev server running at {{DEV_URL}}
+- [ ] Browser open with DevTools ready
+- [ ] Test data prepared (see section 1.3)
+
+### Test 1: [Description]
+**URL:** {{DEV_URL}}/path/to/feature
+**Steps:**
+1. Navigate to the URL above
+2. [Action to perform]
+3. [Expected result to verify]
+
+**Expected Result:** [What should happen]
+**Actual Result:** [ ] PASS / [ ] FAIL - Notes: ___
+
+### Test 2: [Description]
+**URL:** {{DEV_URL}}/path/to/feature
+**Steps:**
+1. [Step 1]
+2. [Step 2]
+
+**Expected Result:** [What should happen]
+**Actual Result:** [ ] PASS / [ ] FAIL - Notes: ___
+
+---
+**Summary:** Complete each test above and mark PASS/FAIL.
+Post results as a comment on this issue.
+```
+
+**Why this matters:**
+- `/test` skill remains useful even without Chrome DevTools MCP
+- Manual testers can follow the structured checklist
+- Test results format remains consistent for reporting
+
 ## Phase 3: Reporting
 
 ### 3.1 Generate Test Summary