candidate-scorecard

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized data access commands were detected in the skill instructions.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a process for handling untrusted interview notes.
  • Ingestion points: Interview notes processed as described in SKILL.md.
  • Boundary markers: None explicitly defined in the prompt instructions.
  • Capability inventory: No tool use, network access, or file system modifications are requested or performed.
  • Sanitization: No specific sanitization or filtering of input data is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 12:04 PM
Security Audit — agent-trust-hub — candidate-scorecard