feishu-doc-writer
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documentation instructs the agent to execute a local shell script located at
./scripts/get_feishu_token.shto retrieve authentication tokens. While this script is intended for legitimate authentication purposes within the skill's environment, executing local scripts involves the execution of shell commands. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest Markdown content from external or user-provided sources and process it for insertion into Feishu documents.
- Ingestion points: The
contentfield in thePOST /documents/{document_id}/convertendpoint and various text elements within the Block JSON structures (e.g.,text_run.content). - Boundary markers: The instructions do not specify any boundary markers or delimiters to isolate untrusted input content from agent instructions.
- Capability inventory: The skill possesses the capability to perform authenticated network requests to the Feishu API (
open.feishu.cn) to create, update, and delete document blocks. - Sanitization: There is no mention of sanitization, validation, or escaping logic applied to the input Markdown before it is converted and uploaded to the remote service.
Audit Metadata