reverse-document
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_CONTEXT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes user-provided implementation code and prototypes to generate documentation. Ingestion points: files at user-specified paths (e.g., src/, Assets/). Boundary markers: the skill explicitly instructs the agent to mark generated content with a provenance banner and use INTENT UNKNOWN for inferred data. Capability inventory: Read, Glob, Grep, Write, Edit, and Bash tools. Sanitization: relies on agent analysis and user approval before final writing. This analysis process could be influenced by malicious instructions embedded in the code comments or logic of the analyzed files.
- [DYNAMIC_CONTEXT_INJECTION]: The skill uses the exclamation-backtick syntax to execute a local configuration script (yaml-helper.sh) during skill load to resolve workflow settings. This is a platform-specific feature used for legitimate configuration management.
- [COMMAND_EXECUTION]: The skill uses the Bash tool to run internal helper scripts (yaml-helper.sh and game-read.sh) for parsing project configuration and game assets. These operations are limited to the skill's own scripts and the user-specified codebase.
Audit Metadata