webman-development
Pass
Audited by Gen Agent Trust Hub on Aug 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill functions by analyzing and modifying project-specific PHP and configuration files, which creates a surface for indirect prompt injection from untrusted source code. \n
- Ingestion points: The agent is instructed to read various codebase files including composer.json, composer.lock, and PHP scripts within the app/ and config/ directories. \n
- Capability inventory: The skill has the capability to write to the local file system, execute shell commands (e.g., php start.php, composer), and perform database operations. \n
- Sanitization: The skill provides extensive documentation (specifically in references/validation-recipes.md) directing the agent to implement input validation and sanitization using the webman/validation component. \n
- Boundary markers: The skill does not implement explicit AI prompt delimiters for the data it processes. \n- [SAFE]: All external references point to official documentation and repositories associated with the framework vendor (e.g., workerman.net, webman.workerman.net). \n- [SAFE]: The skill provides security-conscious instructions, specifically advising against high-risk activities such as piping remote scripts to the shell and hardcoding sensitive credentials in source code.
Audit Metadata