NYC

baseline-restorer

Pass

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • PROMPT_INJECTION (LOW): The skill is susceptible to indirect prompt injection because it processes untrusted data from git logs and file diffs without isolation.
  • Ingestion points: Data enters through git show, git diff, and file reading.
  • Boundary markers: Absent; no delimiters are defined to separate code content from instructions.
  • Capability inventory: Access to Bash and Edit tools allows for potential system impact if the agent is misled.
  • Sanitization: Absent; no filtering is applied to ingested repository data.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 18, 2026, 03:04 AM