discussion-security

# discussion-security

Security specialist participant for discussions

## Installation

This is a CmdForge tool. Install it via:

```bash
cmdforge install discussion-security
```

Or from a collection:

```bash
cmdforge collection install <collection-name>
```

## Usage

This tool is designed to be used in AI discussion workflows, typically called by a discussion orchestrator. It processes discussion context and provides security-focused analysis.

### Basic Usage

```bash
echo "Discussion context about authentication implementation..." | discussion-security
```

### With Callout

```bash
echo "Should we use JWT or session tokens?" | discussion-security --callout "@security what are the security implications?"
```

### Custom Paths

```bash
cat discussion.txt | discussion-security --templates-dir ./my-templates --diagrams-dir ./output --log-file progress.log
```

## Arguments

| Flag | Variable | Default | Description |
|------|----------|---------|-------------|
| `--callout` | `callout` | `""` | Specific question or @mention context |
| `--templates-dir` | `templates_dir` | `templates` | Path to templates directory |
| `--diagrams-dir` | `diagrams_dir` | `diagrams` | Path to save diagrams |
| `--log-file` | `log_file` | `""` | Path to log file for progress updates |

## How It Works

The tool executes a 9-step pipeline:

1. **Phase Detection**: Analyzes the discussion context to determine the current phase
2. **Diagram Preparation**: Sets up diagram path if needed
3. **Progress Logging**: Logs initial progress (if log file specified)
4. **AI Analysis**: Calls Claude Sonnet with security-focused prompt to analyze the discussion
5. **Progress Update**: Logs analysis completion
6. **JSON Extraction**: Parses the AI response for structured output
7. **Response Parsing**: Extracts comment, vote, and optional diagram content
8. **Diagram Saving**: Saves Mermaid diagrams to the specified directory
9. **Final Assembly**: Formats the final response with comment, vote, and diagram reference

The tool specializes in security analysis, examining discussions for:
- Security vulnerabilities and risks
- Authentication and authorization concerns
- Data protection requirements
- Attack vectors and threat modeling
- Security best practices and compliance

## Dependencies

None

## Output Format

The tool outputs JSON with:
- `comment`: Security analysis and recommendations
- `vote`: Security assessment (-1, 0, or 1)
- `diagram`: Optional Mermaid diagram filename (if generated)