dsh-translate-docs
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The workflow relies on executing local repository scripts via
pnpm(such asgen-translation-briefandverify-translation-pairing) to automate diff analysis, structural validation, and consistency recording.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes repository documentation which serves as a surface for indirect prompt injection from untrusted data.\n - Ingestion points: Markdown source files and generated translation briefings are read into the agent context.\n
- Boundary markers: The instructions focus on structural fidelity and terminology but do not define explicit delimiters or instructions to prevent the agent from obeying commands embedded within the documentation text.\n
- Capability inventory: The agent can execute local scripts and perform filesystem writes to record translation pairings.\n
- Sanitization: The skill emphasizes terminology alignment and structural checks but does not specify input sanitization techniques.
Audit Metadata