Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| ADVERSARY_LLM_MODEL | No | LLM model to use | |
| ADVERSARY_LOG_LEVEL | No | Log level for the application | INFO |
| ADVERSARY_LLM_PROVIDER | No | LLM provider (e.g., openai, anthropic) | |
| ADVERSARY_CACHE_SIZE_MB | No | Cache size in megabytes | 200 |
| ADVERSARY_WORKSPACE_ROOT | No | Path to the project workspace root | |
| ADVERSARY_MAX_CONCURRENT_SCANS | No | Maximum number of concurrent scans | 8 |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| adv_scan_code | Scan source code for security vulnerabilities. Results are saved as .adversary.json or .adversary.md in the specified directory. |
| adv_scan_file | Scan a file for security vulnerabilities. Results are saved in the same directory as the target file. |
| adv_scan_folder | Scan a directory for security vulnerabilities. Results are saved in the target directory. |
| adv_diff_scan | Scan security vulnerabilities in git diff changes between branches. Results are saved in the repository root. |
| adv_get_status | Get server status and configuration |
| adv_get_version | Get version information |
| adv_clear_cache | Clear all local cache and data storage |
| adv_mark_false_positive | Mark a finding as a false positive in the .adversary.json file |
| adv_unmark_false_positive | Remove false positive marking from a finding in the .adversary.json file |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |