openspec-to-prompts

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • Prompt Injection (LOW): Vulnerable to Indirect Prompt Injection (Category 8).
  • Ingestion points: Reads specification files including tasks.md, proposal.md, and various spec.md files from the openspec/changes/{change-id}/ directory tree.
  • Boundary markers: Absent. The skill does not define specific delimiters or instructional guards to isolate the data read from files from the agent's logic.
  • Capability inventory: Performs multiple file-system reads and writes, and invokes several sub-skills (prd-lite, prd-clarifier, prd-to-ux, ux-spec-to-prompts) to process and generate content.
  • Sanitization: Absent. The skill logic does not mention escaping or validating the content of the specification files before using them to drive the design pipeline.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:42 PM