deslop

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill defines a workflow for processing external code diffs, which creates a surface for indirect prompt injection.\n
  • Ingestion points: The agent is instructed to "Check the diff against the base branch" as defined in SKILL.md.\n
  • Boundary markers: Absent. The instructions do not include specific delimiters or warnings to ignore adversarial instructions embedded in the code being analyzed.\n
  • Capability inventory: The skill implies file-writing capabilities as it tasks the agent with removing code slop and simplifying structure.\n
  • Sanitization: Absent. The skill does not provide methods for filtering or sanitizing comments or code strings that could contain hidden instructions.\n- [NO_CODE]: The skill does not include any scripts, binaries, or executable files, which significantly reduces the direct attack surface.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 12:27 AM
Security Audit — agent-trust-hub — deslop