direct-response-funnel-copy
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill is comprised of Markdown documentation and reference guides for copywriting. It does not include any Python scripts, Node.js packages, shell commands, or other executable components.\n- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent ingests untrusted external data to produce marketing assets, creating a potential surface for indirect prompt injection.\n
- Ingestion points: The skill requires inputs such as client workshops, briefs, testimonials, and conversion data exports as documented in the Required Inputs table of SKILL.md.\n
- Boundary markers: The instructions lack specific delimiters or isolation markers (e.g., XML tags or clear block quotes) to differentiate user-provided data from system instructions, nor do they include warnings for the agent to ignore instructions embedded within the inputs.\n
- Capability inventory: The skill is primarily instructional and specifies a 'Read and search only' boundary, which limits the potential impact of an injection to the generated text output rather than system-level actions.\n
- Sanitization: No sanitization or validation protocols are defined for the ingested marketing artefacts.
Audit Metadata