ds-lessons
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data sources to generate automated enforcement rules.
- Ingestion points: The skill reads session history, recent commits, code review comments, and agent instruction files (SKILL.md, Step 1).
- Boundary markers: There are no explicit instructions for the agent to ignore or escape instructions embedded in the analyzed data sources.
- Capability inventory: The agent is instructed to modify the codebase (SKILL.md, Step 3) and execute code to verify checks (SKILL.md, Step 4).
- Sanitization: No sanitization or validation of the input data is performed before it is used to define new rules or code checks.
- [DYNAMIC_EXECUTION]: The skill involves the generation and execution of scripts as part of its core workflow.
- Step 4 instructs the agent to create checks (such as tests or lint scripts) and execute them to verify their effectiveness against real-world mistakes.
Audit Metadata