conversation-html-recap
Pass
Audited by Gen Agent Trust Hub on Apr 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a utility for documentation and session summary. It does not attempt to access sensitive system files, hardcode credentials, or execute arbitrary code. The use of a local, self-contained HTML template ensures that the resulting recap remains private and functional without an internet connection.
- [PROMPT_INJECTION]: Evaluation of the indirect prompt injection surface (Category 8) identifies the following characteristics: 1. Ingestion points: Conversation logs and produced artifacts as described in the workflow. 2. Boundary markers: No specific delimiters or safety instructions are defined to ignore potential instructions embedded within the source artifacts. 3. Capability inventory: The skill's operations are limited to reading local context and writing the recap HTML file; it does not utilize network access or shell execution tools. 4. Sanitization: The instructions do not explicitly mandate the escaping of HTML content from artifacts, which is an area for consideration to prevent the rendering of potentially malicious scripts if the generated file is viewed in a browser.
Audit Metadata