Skip to content

oliver-kriska/claude-elixir-phoenix / plugins/elixir-phoenix/skills/full

Deliver a large, cross-domain Phoenix feature or complete end-to-end system through planning, implementation, verification, and review. Use when several coordinated workflows or contexts must ship together. NOT for an existing plan; use /phx:work.

FileKindSize
SKILL.mdskill md9.0 KB
references/cycle-patterns.mdmarkdown4.5 KB
references/example-run.mdmarkdown2.3 KB
references/execution-steps.mdmarkdown4.0 KB
references/safety-recovery.mdmarkdown2.1 KB

references/execution-steps.md

View on GitHub
# Full Cycle Execution Steps

Detailed step-by-step execution for `/phx:full`.

## Step 1: Initialize

```bash
# Create feature slug
FEATURE_SLUG=$(echo "{feature}" | tr '[:upper:]' '[:lower:]' | tr ' ' '-' | tr -cd 'a-z0-9-')

# Create directories
mkdir -p .claude/plans/${FEATURE_SLUG}/{research,reviews,summaries}

# Create feature branch (optional)
git checkout -b feature/$FEATURE_SLUG
```

## Step 2: Discovery Phase

**Purpose**: Gather context and offer user choices before committing to workflow depth.

1. **Quick Codebase Scan** (30-60 seconds):
   - Spawn `phoenix-patterns-analyst` for focused analysis
   - Look for: similar features, related contexts, existing patterns

2. **Assess Complexity**:

   | Score | Level | Recommendation |
   |-------|-------|----------------|
   | <= 2 | LOW | "just do it" -> Skip to WORKING |
   | 3-6 | MEDIUM | "plan it" -> Standard planning |
   | 7-10 | HIGH | "research it" -> Comprehensive planning (4+ agents) |
   | > 10 | CRITICAL | "research it" + security focus |

3. **Present Options**:

   ```
   ## Discovery Summary
   **Feature**: {description}
   **Complexity**: {level}
   **What I Found**: {patterns, contexts}

   **Your options**:
   - "just do it" - Quick implementation
   - "plan it" - Create plan first (standard research)
   - "research it" - Comprehensive plan with deep research
   ```

4. **Route Based on Choice**:
   - "just do it" -> Skip to Step 4 (Work Phase)
   - "plan it" -> Continue to Step 3 (Plan Phase, standard)
   - "research it" -> Continue to Step 3 (Plan Phase, deep)
   - Security features -> Cannot skip planning

**Exit condition**: User selects workflow depth.

## Step 3: Plan Phase

Run `/phx:plan {feature}` (with `--depth deep` for "research it"):

- Spawn research agents (1-2 for standard, 4+ for deep)
- Create phased implementation plan
- Write `.claude/plans/{feature}/plan.md`

**Exit condition**: Plan file exists with checkboxes.

## Step 4: Work Phase (Loop)

Run `/phx:work .claude/plans/{feature}/plan.md`:

```
WHILE unchecked tasks exist:
  1. Find next unchecked task
  2. Route to specialist agent
  3. Execute task
  4. Run verification
  5. IF pass: Mark [x], continue
     IF fail after 3 retries: Create blocker, continue
  6. Log to progress file
```

**Exit condition**: All checkboxes marked OR max retries on blocker.

## Step 5: Review Phase

Run `/phx:review` (append `--codex` when the full run was invoked with
`--codex` — this forwards the flag to the existing `/phx:review --codex`
path, adding the codex-reviewer track):

Spawn 4 parallel review agents (5 with `--codex`):

| Agent | Focus |
|-------|-------|
| elixir-reviewer | Idioms, patterns, code quality |
| testing-reviewer | Test coverage, patterns |
| security-analyzer | Security issues |
| verification-runner | Full test suite |
| codex-reviewer (`--codex` only) | Cross-model second opinion via Codex CLI; SKIPPED if CLI absent |

**Exit condition**: Review complete.

## Step 6: Handle Review Findings

```
IF critical issues found:
  1. Add fix tasks to plan
  2. Go to Step 4 (Work Phase)

IF only warnings:
  1. Log warnings
  2. Continue to completion

IF clean:
  1. Continue to completion
```

## Step 7: Collect Metrics & Complete

Append metrics to progress file:

```markdown
## Metrics

| Metric | Value |
|--------|-------|
| Total Duration | {time} |
| Cycles | {n} |
| Phases | {n} |
| Tasks Completed | {n} |
| Tasks Blocked | {n} |
| Retries | {n} |
| Review Issues Fixed | {n} |
| Files Modified | {n} |
| Tests Added | {n} |
```

Auto-suggest optional follow-ups:

- `/phx:document` for documentation generation
- `/phx:learn-from-fix` to capture lessons learned

Then output completion:

```markdown
## Feature Complete

**Feature**: {feature}
**Duration**: {time}
**Files Modified**: {count}
**Tests Added**: {count}

### Summary

{Brief description of what was implemented}

### Artifacts

- Plan: .claude/plans/{feature}/plan.md
- Progress: .claude/plans/{feature}/progress.md
- Review: .claude/plans/{feature}/reviews/{feature}-review.md

<promise>DONE</promise>
```
planning-and-task-breakdownaddyosmani/agent-skillsBreaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task feels too large to start, when you need to estimate scope, or when parallel work is possible.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
screenpipethedotmack/claude-memSearch screen and audio history recorded by Screenpipe and optionally save selected context into claude-mem. Use when the user asks to recall desktop activity, a meeting, or something seen in another app through Screenpipe.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
contact-widgetnexu-io/open-designSelf-contained floating chat widget with welcome screen, social links, meeting button, and message input. Single HTML file, zero dependencies.Stars: 108kUpdated Claude CodeCodexVerifiedInstall
meeting-notesnexu-io/open-designMeeting notes page — title bar with attendees, agenda checklist, decisions block, action items table with owners + dates, and a "next meeting" footer. Use when the brief mentions "meeting notes", "minutes", "1:1 notes", "all-hands recap", or "会议纪要".Stars: 108kUpdated Claude CodeCodexVerifiedInstall