campaign-plan
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's primary function is generating campaign documents based on brand context. It reads configuration files from application-specific directories (~/.claude-marketing/), which is expected for its stated marketing purpose.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies an ingestion surface where it reads external data from files like profile.json, guidelines, and SOPs. While this data influences the generated output, the skill lacks any dangerous capabilities such as network access, command execution, or file system writing that would allow for an exploitation beyond affecting the plan's text.
- Ingestion points: Configuration files located at
~/.claude-marketing/brands/and~/.claude-marketing/sops/. - Boundary markers: None mentioned; the agent is instructed to read and apply the context directly.
- Capability inventory: None; the skill's workflow is limited to text generation and plan structuring.
- Sanitization: Not specified; the skill relies on the loaded guidelines to define constraints.
Audit Metadata