campaign-brief

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests context from several files within the knowledge/ directory, including personas, market positioning, and competitors. If these files contain malicious instructions, they could influence the agent's planning process or tool calls.
  • Ingestion points: knowledge/icp/personas.md, knowledge/services/, knowledge/markets/positioning.md, knowledge/markets/competitors.md, knowledge/kpis.md, knowledge/learnings.md, and knowledge/company.md.
  • Boundary markers: The skill does not define specific delimiters or instructions to prevent the agent from following directions potentially embedded within the ingested knowledge files.
  • Capability inventory: The skill is configured to write files to the output/campaign-brief/ directory and is designed to trigger other tools like /linkedin-post and /ppt-maker.
  • Sanitization: The skill employs a [UNVERIFIED] tagging system to track data reliability for business metrics, but this does not provide protection against adversarial instructions in the input text.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 06:28 AM
Security Audit — agent-trust-hub — campaign-brief