The Agent Skills Directory

[DATA_EXFILTRATION]: The skill instructs the agent to access and read files from the user's home directory configuration path ~/.claude/plugins/config/.... While intended to migrate existing user preferences, these files may contain sensitive metadata, project settings, or environmental configuration.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting instructions from multiple external files and applying them to the current session context.
Ingestion points: Processes instructions from corporate-legal/CLAUDE.md, ~/.claude/plugins/config/, and various domain-specific SKILL.md files located in the corporate-legal/skills directory.
Boundary markers: The skill lacks explicit delimiters or warnings to the agent to treat external content as data rather than instructions, potentially allowing instructions within those files to override current safety or operational guidelines.
Capability inventory: The agent is granted capabilities for local file system access (read/write), web verification, document rendering, and natural language conversation.
Sanitization: There is no evidence of sanitization, filtering, or validation of the content retrieved from the external source files before it is processed by the agent.
[NO_CODE]: The skill does not bundle any executable scripts or binary files, relying entirely on natural language instructions and existing platform tools.

chinese-legal-corporate