The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs users to install the @wecom/cli package from the npm registry using the command npm install -g @wecom/cli@0.1.8 to enable its core functionality.
[COMMAND_EXECUTION]: The suite utilizes the Bash tool to execute wecom-cli commands across various business domains, including contacts, messaging, and document management. It also recommends using system tools like curl or fetch for certain integration tasks.
[DATA_EXFILTRATION]: The skill defines a fallback mechanism for smartsheet operations that permits sending data to external Webhook URLs provided by the user during the session. While documented for interoperability, this enables a network communication path to arbitrary external endpoints.
[PROMPT_INJECTION]: The skill processes untrusted content from the WeCom platform, creating a surface for indirect prompt injection.
Ingestion points: External data enters the agent context via chat message retrieval (references/wecom-msg-get-message.md), document content fetching (references/wecom-doc-get-doc-content.md), and smartsheet record queries (references/wecom-doc-smartsheet-get-records.md).
Boundary markers: Absent. The instructions do not define specific delimiters or instructions to treat data from these sources as untrusted content.
Capability inventory: The agent has access to the Bash tool, allowing for the execution of CLI tools and network operations.
Sanitization: Absent. There is no mention of filtering, validation, or escaping logic applied to the data retrieved from WeCom before it is processed by the agent.

wecom-unified