easysdd-learning

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a local utility script python easysdd/tools/search-yaml.py to query and filter document metadata. This is used for finding existing documents and preventing duplicates, representing a legitimate functional dependency within the project's own tools.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes existing documentation from the local directory to update or supersede entries, creating a surface for potential indirect instructions to enter the agent's context.\n
  • Ingestion points: Reads existing markdown documents from the easysdd/compound/ directory during Phase 1.5.\n
  • Boundary markers: No specific delimiters or instructions to ignore embedded instructions are defined for the reading phase.\n
  • Capability inventory: The skill's primary capability is writing markdown files to the local easysdd/compound/ directory.\n
  • Sanitization: No explicit sanitization or validation of the ingested document content is described.\n- [DATA_EXFILTRATION]: No network activity, exfiltration patterns, or access to sensitive system directories (e.g., credentials, SSH keys) were identified. The skill's operations are restricted to the local project structure.\n- [SAFE]: The skill incorporates human-in-the-loop checkpoints, requiring the user to review drafts before they are written to disk and to approve modifications to project entry points rather than performing these actions autonomously.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 12:02 PM