adr
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill employs various shell commands (mkdir, rsync, cp, test, grep) to organize the project's documentation structure and ensure the decision index is kept up to date. These operations are restricted to the local filesystem.
- [DYNAMIC_EXECUTION]: The skill performs runtime calculations using Python scripts passed via command-line arguments and heredocs. These scripts are used for harmless tasks such as determining the next sequential ADR number and calculating follow-up review dates.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data which could contain malicious instructions. (1) Ingestion points: Reads RFC files, Pull Request comments, and existing adr-*.md files. (2) Boundary markers: None identified. (3) Capability inventory: File writing, directory creation, and execution of local Python scripts. (4) Sanitization: No explicit sanitization mentioned.
- [SAFE]: No malicious patterns, such as data exfiltration, credential theft, or remote code downloads, were detected. The use of environment variables for path resolution is standard practice for this type of integration.
Audit Metadata