onboard

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to explore the codebase thoroughly, creating a surface where malicious instructions embedded in project files could influence the agent's behavior.
  • Ingestion points: Source files and documentation within the local codebase are read during the exploration phase (SKILL.md).
  • Boundary markers: The instructions lack specific delimiters or "ignore embedded instructions" directives to separate task context from the content of the files being read.
  • Capability inventory: The agent utilizes standard file system access to read project files and write the resulting summary to the .claude/tasks/ directory.
  • Sanitization: No validation or filtering is applied to the content retrieved from the codebase before it is incorporated into the persistent onboarding documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 03:12 AM
Security Audit — agent-trust-hub — onboard