genre-skill-builder
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses git commands to manage the state of the analysis project across different phases.
- [COMMAND_EXECUTION]: A Python validation script is provided in the documentation to check the integrity of the output files.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the analysis of a research corpus.
- Ingestion points: Article sections in markdown or PDF format are read from a user-provided directory.
- Boundary markers: No delimiters are present to separate untrusted document content from agent instructions.
- Capability inventory: The skill can write to the local filesystem, execute shell commands, and generate new agent configuration and instruction files.
- Sanitization: Input content is not filtered for potential instruction-based attacks.
Audit Metadata