The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides a workflow for sending messages where user-supplied content is interpolated into Feishu API calls, creating a potential surface for indirect prompt injection.
Ingestion points: User-provided values for recipient names and message content are ingested via natural language instructions described in SKILL.md.
Boundary markers: The instructions lack explicit delimiters or "ignore instructions" warnings to prevent the agent from being influenced by malicious content embedded in the message body.
Capability inventory: The skill has the capability to send various message types (text, post, interactive) via mcp__feishu__im_v1_message_create and references future capabilities for document and bitable creation.
Sanitization: No specific sanitization, validation, or escaping mechanisms are described for the external content before it is processed by the messaging tools.
[EXTERNAL_DOWNLOADS]: The skill references the official Feishu open platform (open.feishu.cn) through links used to manage application permissions and access developer resources. These references target a well-known service as part of the intended integration workflow.

feishu-messaging