pre-implementation-check
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides shell script templates for dev server health checks and dependency verification. These utilize common commands like 'lsof', 'curl', 'grep', 'test', and 'stat' to inspect the local environment and project files.
- [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface (Category 8). 1. Ingestion points: The agent is instructed to read contents from 'package.json', 'vite.config.ts', 'tasks/progress.txt', and the browser's JavaScript environment using 'agent-browser eval'. 2. Boundary markers: The skill lacks explicit boundary markers or safety instructions to distinguish ingested data from internal agent logic. 3. Capability inventory: The workflow involves file system access ('grep', 'ls', 'stat'), network requests ('curl'), and browser execution ('agent-browser'). 4. Sanitization: There is no evidence of sanitization or validation of the retrieved data (such as version strings or browser variables) before it is used in decision-making processes.
Audit Metadata