The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill creates an attack surface by instructing the agent to process data from external, potentially untrusted sources like Figma design nodes and remote documentation via MCP tools.
Ingestion points: get_design_context (Figma), get_screenshot (Figma), and get-docs (Pixel documentation) in SKILL.md.
Boundary markers: Absent. The instructions do not define delimiters or specific warnings to ignore instructions embedded within the Figma data or documentation strings.
Capability inventory: The agent has the capability to generate and structure Vue 3/Nuxt code, which could be manipulated if the input data contains malicious instructions.
Sanitization: Absent. There are no guidelines for the agent to validate or sanitize the content retrieved from external tools before generating code.
[Data Exposure & Exfiltration] (SAFE): The skill does not attempt to access sensitive local files (like SSH keys or AWS credentials) or perform unauthorized network requests to exfiltrate data.
[Remote Code Execution] (SAFE): No patterns of downloading and executing arbitrary scripts (e.g., curl | bash) were detected. The references to packages are standard for Vue/Nuxt development imports.
[Obfuscation] (SAFE): No Base64, zero-width characters, or other encoding techniques intended to hide malicious intent were found in the provided files.

pixel