# Headless Workflow

**Last Updated**: 2025-11-15

CCS delegation uses Claude Code headless mode with enhanced features for token optimization.

## Core Concept

CCS delegation executes tasks via alternative models using enhanced Claude Code headless mode with stream-JSON output, session management, and cost tracking.

**Actual Command:**
```bash
ccs {profile} -p "prompt"
```

Internally executes:
```bash
claude -p "prompt" --settings ~/.ccs/{profile}.settings.json --output-format stream-json --permission-mode acceptEdits
```

**Docs:** https://code.claude.com/docs/en/headless.md

## How It Works

**Workflow:**
1. User: `/ccs "task"` in Claude Code session (auto-selects profile)
2. CCS detects `-p` flag and routes to HeadlessExecutor
3. HeadlessExecutor spawns: `claude -p "task" --settings ~/.ccs/[selected].settings.json --output-format stream-json --permission-mode acceptEdits`
4. Claude Code runs headless with selected profile + enhanced flags
5. Returns stream-JSON with session_id, cost, turns
6. Real-time tool use visibility in TTY
7. ResultFormatter displays formatted results with metadata

**Enhanced Features:**
- Stream-JSON output parsing (`--output-format stream-json`)
- Real-time tool use visibility (e.g., `[Tool Use: Bash]`)
- Session persistence (`~/.ccs/delegation-sessions.json`)
- Cost tracking (displays USD cost per execution)
- Time-based limits (10 min default timeout with graceful termination)
- Multi-turn session management (resume via session_id)
- Formatted ASCII box output

## Profile Settings

**Location:** `~/.ccs/{profile}.settings.json`

**Examples:**
- GLM: `~/.ccs/glm.settings.json`
- Kimi: `~/.ccs/kimi.settings.json`

**Example content:**
```json
{
  "env": {
    "ANTHROPIC_BASE_URL": "https://api.z.ai/api/anthropic",
    "ANTHROPIC_AUTH_TOKEN": "your-glm-api-key",
    "ANTHROPIC_MODEL": "glm-4.6"
  }
}
```

## Output Format

**Stream-JSON Mode** (automatically enabled):
Each message is a separate JSON object (jsonl format):
```json
{"type":"init","session_id":"abc123def456"}
{"type":"user","message":{"role":"user","content":"Task description"}}
{"type":"assistant","message":{"role":"assistant","content":[{"type":"tool_use","name":"Bash"}]}}
{"type":"result","subtype":"success","total_cost_usd":0.0025,"num_turns":3,"session_id":"abc123def456","result":"Task completed"}
```

**Real-time Progress** (TTY only):
```
[i] Delegating to GLM-4.6...
[Tool Use: Write]
[Tool Use: Write]
[Tool Use: Bash]
[i] Execution completed in 1.5s
```

**Formatted Output** (displayed to user):
```
╔══════════════════════════════════════════════════════╗
║ Working Directory: /path/to/project                 ║
║ Model: GLM-4.6                                       ║
║ Duration: 1.5s                                       ║
║ Exit Code: 0                                         ║
║ Session ID: abc123de                                 ║
║ Cost: $0.0025                                        ║
║ Turns: 3                                             ║
╚══════════════════════════════════════════════════════╝
```

**Extracted Fields:**
- `session_id` - For multi-turn (--resume)
- `total_cost_usd` - Cost per execution
- `num_turns` - Turn count
- `is_error` - Error flag
- `result` - Task output

**Exit codes:** 0 = success, non-zero = error

## Multi-Turn Sessions

**Start session:**
```bash
ccs glm -p "implement feature"
```

**Continue session:**
```bash
ccs glm:continue -p "add tests"
ccs glm:continue -p "run tests"
```

Via slash commands:
```
/ccs "implement feature"           # Auto-selects best profile
/ccs --glm "implement feature"      # Forces GLM profile
/ccs:continue "add tests"           # Continue last session
```

**Session Storage:** `~/.ccs/delegation-sessions.json`

**Metadata:**
- Session ID
- Total cost (aggregated across turns)
- Turn count
- Last turn timestamp
- Working directory

**Expiration:** ~30 days, auto-cleanup

## Usage Patterns

**Single execution:**
```bash
ccs glm -p "task description"
```

**With options:**
```bash
ccs glm -p "task" --permission-mode plan
```

**Continue session:**
```bash
ccs glm:continue -p "follow-up task"
```

All standard Claude Code headless flags are supported. See: https://code.claude.com/docs/en/headless.md

## Error Handling

**Common errors:**
- `Settings file not found` - Profile not configured (`ccs doctor` to diagnose)
- `Claude CLI not found` - Install Claude Code
- `Invalid API key` - Check profile settings in `~/.ccs/{profile}.settings.json`

**Diagnostics:**
```bash
ccs doctor          # Check configuration
ccs --version       # Show delegation status
```

---

## Related Documentation

**Entry Point**: `../SKILL.md` - Quick start and decision framework
**Decision Guide**: `delegation-guidelines.md` - When to delegate
**Error Recovery**: `troubleshooting.md` - Common issues

**Official Docs**: https://code.claude.com/docs/en/headless.md