design-lead

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it is designed to ingest and process the contents of multiple external project files to formulate design strategies.
  • Ingestion points: The skill reads context from projects/<project>/onboarding.md, projects/<project>/positioning.md, and projects/<project>/discovery.md as specified in the 'Before Starting' and 'Context Gathering' sections of SKILL.md.
  • Boundary markers: Absent. There are no explicit delimiters or instructions telling the agent to ignore potentially malicious commands embedded within the project files.
  • Capability inventory: The skill has access to powerful tools including Write (to save direction documents), WebSearch (for competitive analysis), and Read/Glob/Grep (to access the file system).
  • Sanitization: Absent. The skill does not perform validation or filtering on the content retrieved from the project files before using it to generate output or make decisions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 03:35 AM