experiment-design
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read existing configuration files to match formats and naming conventions.
- Ingestion points: SKILL.md (Step 5: "Read existing configs first to match")
- Boundary markers: Absent. The skill does not define specific delimiters for separating user-provided configuration data from its own instructions.
- Capability inventory: The skill generates shell scripts, batch runners, and configuration stubs (Steps 5 and 6).
- Sanitization: Absent. The skill relies on matching existing project patterns without explicit validation of the source content.
- [DYNAMIC_EXECUTION]: The skill involves the generation of executable content based on user requirements.
- Evidence: Step 6 instructs the agent to "Create a shell script or batch runner matching the project's existing patterns."
- Context: This is a standard functional requirement for an experiment design tool to automate the execution of the planned matrix, and the scripts are generated locally based on known project templates.
Audit Metadata