dsh-prose-standard
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes and edits documentation, comments, and UI strings within a repository, creating a surface for indirect prompt injection from untrusted files. Ingestion points: Markdown files, JSDoc, code comments, and UI locale dictionaries in SKILL.md. Boundary markers: None specified to separate content from instructions. Capability inventory: Authorized to edit repository files and execute local scripts like verify-client-ui-i18n. Sanitization: No explicit validation or sanitization of external content is mentioned.
- [COMMAND_EXECUTION]: The skill workflow involves executing shell commands for searching and verifying repository content. Evidence: Instructions in SKILL.md command the use of ripgrep for finding candidates and the execution of verify-client-ui-i18n and behavior tests to check localized strings.
Audit Metadata