visual-component-finder
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a bundled Python script (
scripts/build-component-index.py) and standard shell commands such asgitandgrepto generate and search a component index. These operations are local and consistent with the skill's purpose of codebase analysis. - [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection because it reads and processes untrusted data from the repository and vision-processed screenshots.
- Ingestion points: The script
scripts/build-component-index.pyreads TypeScript and HTML files from the repository, and the agent processes user-provided screenshots. - Boundary markers: There are no explicit markers or safety instructions in the workflow to differentiate between data and instructions when reading file contents.
- Capability inventory: The skill allows the execution of bash commands and the indexing script, which could be exploited if malicious instructions in source files influence agent behavior.
- Sanitization: No sanitization is performed on the text extracted from codebase files or screenshots before it is used in the agent's context.
Audit Metadata