startup-ceo

Pass

Audited by Gen Agent Trust Hub on May 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface\n
  • Ingestion points: The skill reads multiple local markdown files, including PLAN.md, ROADMAP.md, MILESTONE.md, AI-SPEC.md, and the contents of the .planning/ directory in Phase 1 (SKILL.md).\n
  • Boundary markers: There are no explicit instructions to use XML tags or clear delimiters when passing content from these files to sub-agents or the grill-with-docs skill.\n
  • Capability inventory: The skill utilizes Read, Write, Bash, and Agent tools (allowing sub-agent delegation) throughout its execution phases (SKILL.md).\n
  • Sanitization: No sanitization or validation of the ingested markdown content is described before the data is processed or passed to downstream agents.
Audit Metadata
Risk Level
SAFE
Analyzed
May 17, 2026, 03:33 PM