paw-cra-campaign-orchestration
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes parallel sub-agents (Designer, Video Producer, Strategist) via the Agent tool to process production tracks based on campaign briefs.
- [INDIRECT_PROMPT_INJECTION]: The skill parses untrusted campaign briefs (Stage 01) and generates tasks for other AI agents. This represents a tier 2 attack surface where malicious instructions in a brief could influence the downstream behavior of the sub-agents or the orchestrator's dispatch logic.
- [EXTERNAL_DOWNLOADS]: The skill references specific trusted repositories for guidelines and templates within the PawBytes ecosystem, though it does not explicitly invoke remote
curlorwgetcommands in the provided files.
Audit Metadata