learning-capture

Pass

Audited by Gen Agent Trust Hub on May 3, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns, hardcoded secrets, or unauthorized network operations were detected in the skill instructions or reference files.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes data from an external database and user input.
  • Ingestion points: Reads hypothesis, metrics, and experiment data from the database using the get-project command (SKILL.md).
  • Boundary markers: No delimiters or 'ignore' instructions are present to isolate ingested data from the prompt.
  • Capability inventory: The skill has the ability to update database records and write files to the local file system (SKILL.md).
  • Sanitization: No sanitization or validation of external data is specified before its use in generating reports or updating state.
Audit Metadata
Risk Level
SAFE
Analyzed
May 3, 2026, 06:46 AM
Security Audit — agent-trust-hub — learning-capture