The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructs the agent to read sensitive credentials (app_id and app_secret) from a local configuration file at ~/.openclaw/openclaw.json for integration with the Feishu platform.
[DATA_EXFILTRATION]: The harvested credentials are transmitted to https://open.feishu.cn via curl to obtain access tokens. Although Feishu is a well-known service, the extraction of secrets from local storage and their subsequent network transmission represents a security concern.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted user input into its content generation workflow without protective boundaries.
Ingestion points: User messages are ingested during the anygen task prepare step in SKILL.md.
Boundary markers: No delimiters or safety instructions are present to prevent the agent from following instructions embedded within the user input.
Capability inventory: The skill possesses significant capabilities including content task creation (anygen task create), file system access (cat), and network communication (curl).
Sanitization: Input from user messages is not escaped or validated before being used to generate task prompts.
[COMMAND_EXECUTION]: The workflow relies on the execution of multiple shell-based operations, including the anygen CLI tool, curl for file transfers, and jq for parsing sensitive configuration data.

anygen