project-onboard

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill reads several files from the local project environment, including .claude/CLAUDE.md and the ai-context/ directory, to automate its diagnostic logic. Since these files are under user control, they could contain malicious instructions designed to influence the agent's behavior during the onboarding process.
  • Ingestion points: The skill reads content from CLAUDE.md, ai-context/*.md, and .claude/skills/ to determine the project case.
  • Boundary markers: The instructions do not specify any delimiters or safety markers to isolate the content of these files from the agent's instructional context.
  • Capability inventory: While this specific skill is read-only, it serves as a gateway that recommends the execution of other high-capability skills (e.g., /project-fix, /sdd-apply) which have file-system write permissions.
  • Sanitization: No sanitization or validation logic is defined for the external content read from the file system.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 05:14 AM
Security Audit — agent-trust-hub — project-onboard