performance-review-self-assessment

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE]: The skill accesses local directory names to identify team members and performs file-write operations to save sensitive performance assessments.
  • Evidence: In SKILL.md, the agent is instructed to look up subdirectories in a quarter folder (e.g., Q12026/) to list team members and save a markdown file to <quarter>/<person>/sources/.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests directory names from the local workspace as data, creating a surface where maliciously crafted directory names could influence agent behavior.
  • Ingestion points: Subdirectory names in the quarter folder are used to populate the team member list in SKILL.md (Step 1).
  • Boundary markers: No explicit boundary markers or instructions are provided to the agent to disregard potential instructions embedded in directory names.
  • Capability inventory: The skill has the capability to list directories and write files to the local disk.
  • Sanitization: There is no evidence of sanitization or validation performed on the directory names before they are presented to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 02:32 PM
Security Audit — agent-trust-hub — performance-review-self-assessment