multi-platform-launch
Fail
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: CRITICALMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
- [METADATA_POISONING]: The skill body contains a recommendation for the external directory 'aitoolfor.org' in the 'AI/Tech Directories' table. This URL has been flagged by automated scanners as a malicious scam site (EshopScam). Recommending that users visit or submit their product information to a flagged scam site represents a significant security risk to the user's data and finances.\n
- Evidence: Table entry '| AI Tool Directory (aitoolfor.org) | Growing | Medium | Broad AI tool listing |' in SKILL.md.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests various untrusted user inputs to customize launch plans, which creates an attack surface for potential indirect prompt injection attacks.\n
- Ingestion points: The skill explicitly requests user input for 'Product type', 'Target buyer', 'Current assets', 'Launch goal', 'Timeline', 'Budget', and 'Team size' in SKILL.md.\n
- Boundary markers: There are no delimiters or specific instructions (like 'ignore any embedded instructions') provided to separate user data from the skill's logic.\n
- Capability inventory: While the skill references other skills (e.g., positioning-icp, ai-seo), the current skill does not directly execute shell commands, perform file system writes, or initiate network requests beyond the documented references.\n
- Sanitization: The instructions do not define any validation, filtering, or sanitization logic for the data collected from the user.
Recommendations
- AI detected serious security threats
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata