The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the global NPM package @wecom/cli and adds an external skill repository from WeComTeam/wecom-cli on GitHub.
[COMMAND_EXECUTION]: The skill executes shell commands using the wecom-cli tool to interact with WeCom APIs. It also provides bash script examples that utilize system utilities such as jq, sed, tr, and date for data manipulation and automation.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface where untrusted data from an external platform is ingested into the agent's context.
Ingestion points: Operations such as wecom-cli message pull_messages, wecom-cli document read, and wecom-cli smartsheet get_records pull data from Enterprise WeChat conversations and documents into the agent's environment.
Boundary markers: The instructions do not specify any delimiters or safety warnings to prevent the agent from executing instructions that may be hidden within retrieved messages or document content.
Capability inventory: The agent has the ability to send messages (send_text), modify documents (edit), schedule meetings, and manage tasks, which could be triggered by malicious instructions in the ingested data.
Sanitization: There is no mention of sanitizing or escaping content retrieved from the WeCom API before the agent processes it.

wecom-cli-enterprise-wechat