The Agent Skills Directory

Indirect Prompt Injection (LOW): The skill provides templates for data ingestion that could be leveraged for indirect prompt injection if the resulting application processes untrusted user input without sanitization.
Ingestion points: examples/components/input.md, examples/components/form.md (components like u-input and u-form).
Boundary markers: Absent; the templates do not include specific instructions for the agent to ignore instructions embedded in user-provided data.
Capability inventory: examples/tools/http.md (network requests via request), examples/tools/intro.md (persistent storage via setStorage).
Sanitization: Absent; the skill focuses on UI implementation rather than data validation/sanitization logic.
Unverifiable Dependencies (SAFE): The skill references the uview-pro package. While not on the specific 'Trusted Sources' list provided in the instructions, it is the primary subject of the skill and a widely used library in the uni-app ecosystem. The installation instructions (npm install uview-pro) are standard and expected for this use case.
Data Exposure (SAFE): While examples/tools/http.md demonstrates how to add an 'Authorization' header, it uses a generic placeholder ('Bearer token') and does not contain hardcoded secrets or access to sensitive system paths.

uview-pro-vue3