growth-ceo

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [PROMPT_INJECTION]: The skill uses an extensive persona-based system ("Growth CEO") to provide strategic advice. Additionally, it presents a vulnerability surface for indirect prompt injection.\n
  • Ingestion points: The skill is instructed to read the codebase, README, and project documentation (SKILL.md, Workflow Step 1).\n
  • Boundary markers: Absent; there are no instructions to ignore or isolate commands that may be embedded in the analyzed data.\n
  • Capability inventory: The skill uses Write and Edit tools to persist initiatives as markdown files (SKILL.md, Workflow Step 3).\n
  • Sanitization: Absent; the skill does not filter or validate content ingested from the project files before processing.\n- [COMMAND_EXECUTION]: The skill uses file system tools to perform project analysis and document generation.\n
  • Evidence: The allowed-tools section in SKILL.md includes Read, Write, Edit, Glob, and Grep.\n- [DATA_EXFILTRATION]: The skill accesses sensitive project data including source code and internal documentation.\n
  • Evidence: Workflow Step 1 requires reading the entire codebase. While there is no direct network exfiltration, project secrets or sensitive logic could be inadvertently exposed in the generated initiative files.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 07:31 AM