markstream-vue2-vite
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The workflow instructions attempt to reduce user oversight by granting the agent pre-authorization for routine source and dependency modifications.
- Evidence: The workflow section in SKILL.md states that a 'requested scoped integration already authorizes routine source/dependency edits', instructing the agent to only seek approval for material scope expansions.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to handle and render external Markdown streams, creating a vulnerability surface for indirect prompt injection or cross-site scripting (XSS).
- Ingestion points: Untrusted data enters the context through the 'content' property of the 'MarkdownRender' component (SKILL.md).
- Boundary markers: The skill does not provide explicit delimiters or instructions for the agent to ignore embedded commands in the data stream.
- Capability inventory: The skill enables the rendering of Markdown, HTML, and Mermaid diagrams from external sources.
- Sanitization: The instructions recommend maintaining 'safe' HTML and 'strict' Mermaid rendering configurations to reduce exposure to malicious payloads.
Audit Metadata