antivibe
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes several local shell scripts (analyze-code.sh, capture-phase.sh, find-resources.sh, and generate-deep-dive.sh) to perform file system operations. These scripts identify recently modified files, extract code structure and metadata using grep, and write generated markdown reports to a deep-dive/ directory.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through its core functionality of analyzing project code. Malicious instructions placed inside code comments or strings in the files being analyzed could be interpreted by the explainer agent as instructions.
- Ingestion points: The explainer agent and scripts read content from any file in the current working directory targeted for a deep dive.
- Boundary markers: The skill does not employ delimiters or specific instructions to the agent to distinguish between the code data being analyzed and the task instructions.
- Capability inventory: The agent can execute local shell scripts and write markdown files to the project directory.
- Sanitization: There is no mechanism in place to sanitize or filter the content of the analyzed files before they are passed to the agent for processing.
Audit Metadata