playtest-report
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEDYNAMIC_CONTEXT_INJECTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_CONTEXT_INJECTION]: The skill executes a local shell script (
yaml-helper.sh) at load time using the!prefix. This is used to resolve configuration keys likereview_modeandautomationfor the session. This is a standard project-specific tooling pattern. - [INDIRECT_PROMPT_INJECTION]: The skill ingests raw, untrusted data from playtest notes provided via a file path.
- Ingestion points: Raw notes read from
path-to-notesin SKILL.md. - Boundary markers: Not explicitly defined; the agent processes the file content directly.
- Capability inventory: Includes file system writing via
Writeand sub-agent spawning viaAgenttool. - Sanitization: No explicit sanitization or escaping of the note content is performed before processing.
- [COMMAND_EXECUTION]: The skill uses the Bash tool restricted via frontmatter to a specific local script (
yaml-helper.sh). This script is used for configuration resolution and does not interact with remote resources or sensitive user data.
Audit Metadata