style-guide
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the codebase being analyzed, which could potentially contain malicious instructions intended to influence the analyzer agents.
- Ingestion points: The skill reads files throughout the user's codebase during analysis (via specialized analyzer agents).
- Boundary markers: No specific boundary markers or instructions to ignore embedded instructions are provided in the sub-agent prompts.
- Capability inventory: The skill can spawn parallel agents and write findings to a local file (
.claude/codebase-style.md). - Sanitization: No explicit sanitization or filtering of codebase content is mentioned before synthesis into the final report.
- [COMMAND_EXECUTION]: The skill uses standard file system exploration commands (
ls -la,glob,grep) to detect frameworks and analyze patterns. These operations are restricted to the local codebase and are necessary for the skill's primary function.
Audit Metadata