custom-prompt
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes instructions from stored Markdown files, creating a surface for indirect prompt injection. This is mitigated by mandatory user confirmation gates and explicit instructions that stored prompts cannot waive session safety rules or git discipline. Evidence includes ingestion from local prompt files, boundary markers via user questions, and capabilities for file operations.
- [COMMAND_EXECUTION]: The skill involves reading, writing, and deleting Markdown files within the project's documentation directories for prompt management.
Audit Metadata