cs-onboard

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill establishes a vulnerability surface for indirect prompt injection by capturing user input through interviews and saving it unsanitized into a shared configuration file (~/.claude/company-context.md) intended for use by multiple agents.
  • Ingestion points: Data enters the context through responses to interview probes defined in SKILL.md and references/interview-guide.md.
  • Boundary markers: The output template (templates/company-context-template.md) lacks sufficient delimiters or instructions to prevent downstream agents from misinterpreting captured text as instructions.
  • Capability inventory: The skill performs file-write operations to the user's local directory to persist the context.
  • Sanitization: No validation, escaping, or filtering of the captured founder responses is performed before storage.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:48 PM
Security Audit — agent-trust-hub — cs-onboard