verification-rules
4-layer verification system for task completion. Use when verifying that a task has been fully completed before marking it done.
What this skill does
# Verification Rules Skill
This skill defines the 4-layer verification system used by CEO Ralph.
## The 4 Verification Layers
Every task must pass ALL 4 layers before being marked complete.
```
┌─────────────────────────────────────────────┐
│ Layer 1: Contradiction │
│ Check for conflicting statements │
└─────────────────────────────────────────────┘
│
▼ PASS
┌─────────────────────────────────────────────┐
│ Layer 2: Uncommitted │
│ Verify all changes committed │
└─────────────────────────────────────────────┘
│
▼ PASS
┌─────────────────────────────────────────────┐
│ Layer 3: Checkmark │
│ Verify task marked [x] in tasks.md │
└─────────────────────────────────────────────┘
│
▼ PASS
┌─────────────────────────────────────────────┐
│ Layer 4: Signal │
│ Verify TASK_COMPLETE signal present │
└─────────────────────────────────────────────┘
│
▼ PASS
VERIFIED ✓
```
## Layer 1: Contradiction Detection
**Purpose**: Catch outputs that claim completion but are actually incomplete.
### Detection Patterns
```javascript
const contradictions = [
// Says complete but needs manual work
/requires manual.*TASK_COMPLETE/i,
/manual intervention.*TASK_COMPLETE/i,
/needs human.*TASK_COMPLETE/i,
// Says complete but has TODOs
/TODO.*TASK_COMPLETE/i,
/FIXME.*TASK_COMPLETE/i,
/XXX.*TASK_COMPLETE/i,
// Says complete but incomplete
/not implemented.*TASK_COMPLETE/i,
/placeholder.*TASK_COMPLETE/i,
/skeleton.*TASK_COMPLETE/i,
// Says complete but blocked
/blocked.*TASK_COMPLETE/i,
/cannot complete.*TASK_COMPLETE/i,
];
```
### Verification Process
```markdown
**Layer 1: Contradiction Detection**
Checking output for contradictory statements...
Patterns checked:
- [x] "requires manual" + completion
- [x] "TODO/FIXME" + completion
- [x] "not implemented" + completion
- [x] "blocked" + completion
Result: {PASS/FAIL}
{If FAIL: Found contradiction: "{matched text}"}
```
### On Failure
- Do NOT mark task complete
- Report specific contradiction found
- Retry with instruction to either:
- Complete the implementation, OR
- Remove the TASK_COMPLETE signal and report blocker
## Layer 2: Uncommitted Files Check
**Purpose**: Ensure all changes are committed to git.
### Verification Process
```bash
# Check spec directory for uncommitted changes
git status ./specs/{specName}/ --porcelain
# Check modified files from task
git status {file1} {file2} --porcelain
```
### Expected Results
- Empty output = All committed ✓
- Any output = Uncommitted files ✗
```markdown
**Layer 2: Uncommitted Files Check**
Checking for uncommitted changes...
Files checked:
- ./specs/{specName}/tasks.md
- ./specs/{specName}/.progress.md
- {modified files from task}
Result: {PASS/FAIL}
{If FAIL: Uncommitted files: {list}}
```
### On Failure
- Run git commit for spec files
- Request worker to commit code changes
- Retry verification
## Layer 3: Checkmark Verification
**Purpose**: Ensure task is marked complete in tasks.md.
### Verification Process
```bash
# Check if task line shows [x]
grep -E "^\s*-\s*\[x\]\s*{taskId}" ./specs/{specName}/tasks.md
```
### Expected Results
- Match found = Marked complete ✓
- No match = Not marked ✗
```markdown
**Layer 3: Checkmark Verification**
Checking tasks.md for completion mark...
Looking for: `- [x] {taskId}`
Found: {matched line or "NOT FOUND"}
Result: {PASS/FAIL}
```
### On Failure
- Update tasks.md to mark task `[x]`
- Commit the change
- Re-verify
## Layer 4: Signal Verification
**Purpose**: Ensure explicit completion signal is present.
### Verification Process
```javascript
const output = workerOutput;
const hasSignal = output.includes("TASK_COMPLETE");
```
### Expected Results
- Signal found = Complete ✓
- No signal = Incomplete ✗
```markdown
**Layer 4: Signal Verification**
Checking for completion signal...
Looking for: `TASK_COMPLETE`
Found: {YES/NO}
Result: {PASS/FAIL}
```
### On Failure
- If output looks complete, add signal and pass
- If output is incomplete, retry task
## Combined Verification Report
```markdown
## Verification Report: Task {id}
| Layer | Check | Status |
|-------|-------|--------|
| 1 | Contradiction Detection | {PASS/FAIL} |
| 2 | Uncommitted Files | {PASS/FAIL} |
| 3 | Checkmark in tasks.md | {PASS/FAIL} |
| 4 | Completion Signal | {PASS/FAIL} |
**Overall**: {VERIFIED/FAILED}
```
## Quality Gate Verification
After the 4 layers pass, run quality gates:
```bash
# Discover and run quality commands
npm run lint # or discovered lint command
npm run test # or discovered test command
npm run build # or discovered build command
npm run check-types # if TypeScript
```
### Quality Gate Report
```markdown
## Quality Gates: Task {id}
| Gate | Command | Status | Output |
|------|---------|--------|--------|
| Lint | npm run lint | {PASS/FAIL} | {summary} |
| Test | npm test | {PASS/FAIL} | {summary} |
| Build | npm run build | {PASS/FAIL} | {summary} |
| Types | npm run check-types | {PASS/FAIL} | {summary} |
**Overall**: {PASS/FAIL}
```
## Escalation Triggers
Escalate to user when:
- Same verification layer fails 3 times
- Quality gate fails and fix is unclear
- Contradiction cannot be resolved
- Git state is corrupted
Related in Productivity
gitea-workflow
IncludedOrchestrate agile development workflows for Gitea repositories using the tea CLI. Use when working with Gitea-hosted repos and asking to 'run the workflow', 'continue working', 'what's next', 'complete the task cycle', 'start my day', 'end the sprint', 'implement the next task', or wanting guided step-by-step development assistance. Keywords: workflow, orchestrate, agile, task cycle, sprint, daily, implement, review, PR, standup, retrospective, gitea, tea.
microsoft-graph-gateway
IncludedRoute Microsoft Graph work in this workspace. Use when users want to read or write Outlook mail, calendar events, contacts, OneDrive or SharePoint files, Teams, Planner, To Do, users, groups, directory data, or arbitrary Microsoft Graph endpoints from VS Code. Prefer WorkIQ for common read scenarios. Use Microsoft Graph for write actions and gap-read scenarios that need exact Graph properties, filters, permissions, or endpoints.
copilotkit
IncludedUse when building with CopilotKit — setup, development, integrations, debugging, upgrading, or contributing. Routes to the appropriate specialized skill based on the task.
wordly-wisdom
IncludedProvides calibrated decision analysis using Charlie Munger-style multiple mental models, inversion, incentive mapping, circle-of-competence checks, misjudgment audits, second-order effects, and forecast updates. Use when the user asks for an oracle take, a hard call, a decision memo, a premortem, an outside view, a red-team, a sanity-check, what am I missing, think this through, or wants a strategy, hire, investment, plan, product, partnership, or major life choice analysed. Avoid for simple factual lookups or time-sensitive legal, medical, or market questions without fresh evidence.
swain-session
IncludedSession management and project status dashboard. Owns the full session lifecycle (start/work/close/resume), focus lane, bookmarks, worktree detection, and tab naming. Also serves as the project status dashboard — shows active epics, progress, actionable next steps, blocked items, tasks, GitHub issues, and recommendations. Worktree creation is deferred to swain-do task dispatch (SPEC-195). Triggers on: 'session', 'status', 'what's next', 'dashboard', 'overview', 'where are we', 'what should I work on', 'show me priorities', 'bookmark', 'focus on', 'session info'.
gandi
IncludedComprehensive Gandi domain registrar integration for domain and DNS management. Register and manage domains, create/update/delete DNS records (A, AAAA, CNAME, MX, TXT, SRV, and more), configure email forwarding and aliases, check SSL certificate status, create DNS snapshots for safe rollback, bulk update zone files, and monitor domain expiration. Supports multi-domain management, zone file import/export, and automated DNS backups. Includes both read-only and destructive operations with safety controls.