The Agent Skills Directory

[SAFE]: The skill's instructions establish a standard operational scope for auditing UI code within a project's source directory (e.g., /src). No malicious patterns, such as credential harvesting or unauthorized privilege escalation, were detected.
[PROMPT_INJECTION]: The skill uses natural instructional language to define audit modes and processes. No attempts to override agent behavior, bypass safety filters, or extract system prompts were identified.
[EXTERNAL_DOWNLOADS]: The skill references official Iress packages, including @iress-oss/ids-components@beta and @iress-oss/ids-tokens@beta. These are legitimate vendor-owned resources. Additionally, external URLs point to reputable sources such as Nielsen Norman Group and Chromatic (a well-known service for Storybook hosting).
[DATA_EXFILTRATION]: While the skill reads project source files and configuration (package.json), this is required for its auditing function. No unauthorized network operations or data exfiltration patterns were observed.

ui-doctor