planning-campaigns

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill functions as an orchestrator, coordinating specialized sub-skills for research and production. It maintains project data in a dedicated directory (./supercmo-files/campaigns/) and manages the workflow through a sequence of approved steps. File operations and network usage are consistent with the skill's purpose as a marketing tool.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for indirect prompt injection by guiding the agent to ingest external marketing and product data. 1. Ingestion points: Product URLs, brand websites, and competitor advertising data are ingested during the research phase (SKILL.md and finding-concepts.md). 2. Boundary markers: The instructions do not define specific delimiters for separating system prompts from external research. 3. Capability inventory: The skill triggers production sub-skills for ad generation and writes creative concepts to local files (SKILL.md Step 2, 3, and 6). 4. Sanitization: No specific data sanitization or validation logic is specified for the ingested external content. This surface is necessary for the skill's creative analysis functions and does not represent a malicious pattern.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 11:11 AM
Security Audit — agent-trust-hub — planning-campaigns