copywriting
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from a local file (.claude/product-marketing-context.md) and user-provided copy. However, the skill lacks exploitable capabilities such as shell access, network requests, or file system modifications, rendering this ingestion surface safe.
- Ingestion points: Reads context from .claude/product-marketing-context.md and processes user inputs for copywriting tasks.
- Boundary markers: No explicit markers are defined to isolate untrusted input within the prompt instructions.
- Capability inventory: No subprocess calls, network requests (curl/wget), or file-write operations are present in any of the skill's files.
- Sanitization: No input sanitization or filtering is performed as the skill's purpose is creative text generation.
- [SAFE]: The skill and its associated reference files (copy-frameworks.md, natural-transitions.md) contain purely static markdown content and instructional text. It does not include any scripts or automation that could compromise the host environment.
Audit Metadata