product-launch-marketer
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [SAFE]: The skill's functionality is delivered entirely through static markdown files and configuration metadata. There are no scripts (Python, JavaScript, shell), binaries, or dynamic commands included in the package.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that ingests untrusted user input (product descriptions, launch objectives) which is then processed by the agent to generate marketing plans.
- Ingestion points: User-provided arguments for 'product', 'audience', and 'business objective' in SKILL.md.
- Boundary markers: The instructions lack explicit delimiters (e.g., XML tags) or 'ignore embedded instructions' warnings for user-supplied data.
- Capability inventory: The skill is limited to 'Read' and 'Write' tools for file system interaction.
- Sanitization: No sanitization or validation of external input is implemented within the instructions, relying on the model's base safety guardrails.
- [NO_CODE]: No executable code was detected in any of the provided files. All logic is expressed as natural language instructions for the AI agent.
Audit Metadata