create-adr
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for documentation and architectural record-keeping. Analysis of the instructions and templates confirms they are limited to benign text generation and standard file system navigation to maintain sequential document numbering.
- [DATA_EXFILTRATION]: No network operations or attempts to send data to external servers were found. The skill only interacts with the local file system to read filenames and save new documentation.
- [PROMPT_INJECTION]: The skill contains instructional guidance for the agent to maintain formatting and language consistency, but lacks any patterns associated with bypassing safety guidelines or overriding system prompts.
- [SAFE]: While the skill ingests user input and reads local filenames, representing a surface for indirect prompt injection, the risk is minimal as the skill's capabilities are restricted to document generation and lacks dangerous execution or network operations. Evidence: 1. Ingestion points: user messages and local file system paths; 2. Boundary markers: absent; 3. Capability inventory: file system read and write; 4. Sanitization: markdown template structure.
Audit Metadata