The Agent Skills Directory

[DATA_EXPOSURE]: The skill reads the authentication credentials (appId and appSecret) from the local configuration file located at ~/.openclaw-autoclaw/openclaw.json to generate a tenant_access_token.- [EXTERNAL_DOWNLOADS]: The skill makes network requests to the official Feishu API domain https://open.feishu.cn for authentication and file uploading. As Feishu is a well-known service, these operations are considered standard for the intended integration.- [INDIRECT_PROMPT_INJECTION]: The skill provides a mechanism to read files from the local filesystem and transmit them to an external service, creating a surface for potential data exfiltration if the agent is instructed to process sensitive files.
Ingestion points: The skill accepts a file path as input (e.g., /path/to/your/file) in the upload step.
Boundary markers: Absent; the code does not include specific delimiters or warnings to ignore instructions embedded within the files being processed.
Capability inventory: The skill utilizes open() for file reading and requests.post() for network operations.
Sanitization: Absent; the skill reads and uploads file content directly without validation or filtering.

feishu-send-file