collect-task-context

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists entirely of markdown instructions and templates for organizing task-specific information.
  • [DATA_EXPOSURE]: The skill includes explicit instructions in the 'Exclude' section to omit secrets, tokens, and private credentials, demonstrating a security-conscious design.
  • [INDIRECT_PROMPT_INJECTION]: While the skill involves reading project documentation and files (ingestion points) to prepare a handoff packet, it lacks any active capabilities such as subprocess execution, network requests, or file-writing tools. The resulting packet is purely text-based, posing no significant risk of indirect prompt injection affecting the agent's operating environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 07:05 AM
Security Audit — agent-trust-hub — collect-task-context