founder-mode

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user data via the <question> parameter which is then used to drive routing decisions and generate file-based summaries.
  • Ingestion points: The <question> input provided by the user in the /cs:founder-mode command (SKILL.md).
  • Boundary markers: Absent. The routing logic directly parses the user's question for signals without explicit delimiters or instructions to ignore embedded commands.
  • Capability inventory: The skill has the capability to invoke multiple other agents (e.g., cs-cfo-advisor, cs-ceo-advisor) and performs file-write operations to the ~/.claude/briefs/ directory.
  • Sanitization: There is no evidence of input sanitization, escaping, or validation of the user-provided question before it is processed by the routing logic or written to a brief file.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 06:08 PM
Security Audit — agent-trust-hub — founder-mode