exec

# Exec Skill: Autonomous Task Execution

## Purpose

Execute tasks autonomously with automatic library discovery, 3-tier functional validation, and atomic git commits. Provides a complete autonomous execution workflow that:
1. Discovers existing libraries (don't reinvent wheels)
2. Validates changes functionally (not just compilation)
3. Commits atomically (only validated code enters git history)
4. Retries on failure (with rollback and research)

**Key Innovation**: Combines Shannon Framework's multi-agent orchestration (/shannon:wave) with Shannon CLI's validation and git automation for truly autonomous, validated execution.

---

## When to Use This Skill

### Primary Use Cases

**MANDATORY when user says**:
- "Execute this task autonomously"
- "Add [feature] with validation and commit"
- "Build [component] using existing libraries"
- "Implement [functionality] and ensure it works"

**RECOMMENDED when**:
- Task requires code generation AND validation
- Library discovery would save time
- Git automation desired
- Quality gates needed before commit

**DO NOT USE when**:
- Just analyzing (use spec-analysis instead)
- Just planning (use phase-planning instead)
- Just generating (use wave-orchestration directly)
- Manual validation preferred

---

## Workflow

### Phase 1: Context Preparation (30-60s)

**Action**: Invoke /shannon:prime for session setup

```
@skill context-preservation
  operation: prepare
  task_focused: true
```

**Purpose**:
- Discover available skills and tools
- Verify MCP connections (Serena required)
- Load project context
- Restore any previous session state

**Output**: Session ready, skills discovered, context loaded

### Phase 2: Library Discovery (5-30s)

**Action**: Call Shannon CLI library discoverer

```bash
# Execute CLI command to search registries
shannon discover-libs "[feature extracted from task]" --category [ui|auth|networking|data|forms] --json
```

**Purpose**:
- Search package registries (npm, PyPI, CocoaPods, Maven, crates.io