ds-full-pipeline
Pass
Audited by Gen Agent Trust Hub on May 6, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill coordinates research tasks using modular commands. It includes mandatory manual checkpoints ('Gate 1' and 'Gate 2') which require user approval before the agent proceeds with automated experiments or writing.- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external research data and user-provided topics.
- Ingestion points: User-provided research problems via
$ARGUMENTSand literature survey results during the 'Scout' stage. - Boundary markers: Explicit delimiters for external data are absent in the orchestration instructions.
- Capability inventory: The pipeline triggers experimental code execution (
/ds-experiment), automated paper drafting (/ds-write), and repository persistence (Git as memory). - Sanitization: No specific sanitization or filtering of research content is described in the pipeline instructions.
Audit Metadata