mission-planner

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through its automated context gathering logic.
  • Ingestion points: Phase 1 instructions direct the agent to parse 'CLAUDE.md' and scan existing project files to understand constraints and preferences.
  • Boundary markers: The instructions lack requirements for delimiters or warnings to ignore embedded instructions within processed project files.
  • Capability inventory: The skill performs file system write operations to '.claude/agents/', '.claude/skills/', and '.claude/teams/', appends to 'usage-log.jsonl', and can invoke the 'Agent Creator' skill to generate new agent instructions.
  • Sanitization: No sanitization or validation is applied to the data ingested from project files before it is interpolated into the reasoning process for team design.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 09:48 PM
Security Audit — agent-trust-hub — mission-planner