debian-linux-triage

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted user input to generate system administration commands, creating a surface for indirect prompt injection.
  • Ingestion points: The skill accepts variables ${input:ProblemSummary} and ${input:Constraints} which ingest data from potentially untrusted sources.
  • Boundary markers: There are no boundary markers or instructions to ignore embedded commands within the user-provided input strings.
  • Capability inventory: While the skill does not execute code itself, its primary purpose is to generate high-privilege commands (e.g., using apt, systemctl, dpkg) for the user to run.
  • Sanitization: There is no evidence of input sanitization or validation to filter out malicious instructions hidden within the problem description.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 05:25 AM