The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the global NPM package @wecom/cli from the public registry and downloads an external agent skill repository from GitHub using npx skills add WeComTeam/wecom-cli.
[COMMAND_EXECUTION]: The skill operates by executing the wecom-cli command-line utility for all interactions with the WeCom platform, including file operations and data management.
[DATA_EXFILTRATION]: The skill provides capabilities to retrieve sensitive enterprise data, such as message history (pull_messages), document content (document read), and contact lists (get_userlist), making this information available to the AI agent.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it ingests untrusted content from external messages and documents while possessing broad modification capabilities within the same environment.
Ingestion points: Data from the WeCom environment enters the agent context via wecom-cli message pull_messages, wecom-cli document read, and wecom-cli smartsheet get_records as described in SKILL.md.
Boundary markers: The skill does not define specific delimiters or instructional warnings to prevent the agent from obeying commands embedded in the retrieved WeCom content.
Capability inventory: The skill allows the agent to execute subprocess commands for creating and editing messages, documents, tasks, and meetings.
Sanitization: There is no mention of sanitization or validation of the content pulled from the WeCom API before it is processed by the agent.

wecom-cli-enterprise-wechat