x-create
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill processes user-provided content from the
assets/templates/directory to learn writing styles. This creates a surface where malicious instructions could be embedded in those template files. - Ingestion points: Files located in
assets/templates/high-value/,sharp-opinion/,trending-comment/,story-insight/, andtech-analysis/(read inSKILL.mdStep 1). - Boundary markers: None identified. The instructions direct the agent to "Learn style from examples" without specifying delimiters or warnings to ignore embedded commands.
- Capability inventory: Primarily text generation and context loading. Mentions integration with an
/x-publishcommand (likely a separate skill). - Sanitization: None identified. The skill directly incorporates learned patterns into the generation workflow.
Audit Metadata