marketing-plan
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an ingestion surface for untrusted data that could influence the agent's behavior during the marketing plan generation process.
- Ingestion points: In Phase 1 (Step 1.2), the skill reads all available materials provided by the client (such as decks, audits, and Slack notes) from the
~/marketing-plans/{client-slug}/materials/directory. - Boundary markers: The instructions do not define clear delimiters or "ignore embedded instructions" warnings when processing these external files, increasing the risk that malicious instructions hidden within them could be obeyed.
- Capability inventory: The skill possesses significant capabilities across its scripts, including writing files to the local file system and accessing numerous external services via MCP integrations (Stripe, Ahrefs, GitHub, GA4, Customer.io).
- Sanitization: There is no documented mechanism for sanitizing or validating the content of the ingested materials before they are interpreted by the model.
Audit Metadata