blueprint

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted user input from task descriptions to generate technical plans and context snapshots. This creates an indirect prompt injection surface where malicious instructions embedded in a task brief could influence the planning logic, architectural decisions, or the behavior of subsequent execution agents such as $forge or $team.
  • Ingestion points: The skill accepts arbitrary strings via the $blueprint command and the Consensus Workflow.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the user-provided tasks are defined.
  • Capability inventory: The skill performs file writes to the .rcs/context/ directory and orchestrates the launch of other high-privilege execution skills.
  • Sanitization: No sanitization or validation of the input task description is performed before it is interpolated into the planning prompt or written to the file system.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 11:36 AM
Security Audit — agent-trust-hub — blueprint