open-spec-apply

Pass

Audited by Gen Agent Trust Hub on Apr 19, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing external task descriptions and design artifacts to drive code modifications.
  • Ingestion points: Reads content from 'specs/changes//' including proposal.md, design.md, and tasks.md (SKILL.md).
  • Boundary markers: Absent. The instructions do not specify any delimiters or safety warnings to distinguish between task data and agent instructions.
  • Capability inventory: The agent is authorized to 'Make the code changes required' (SKILL.md), which involves broad file-writing capabilities across the codebase.
  • Sanitization: Absent. There is no mention of validating or escaping the content read from the specification files before using it to guide implementation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 19, 2026, 11:09 AM
Security Audit — agent-trust-hub — open-spec-apply