The Agent Skills Directory

[COMMAND_EXECUTION]: The skill extensively uses child_process modules (execSync, spawn, spawnSync) to manage system-level operations. This includes executing Git commands for code synchronization and interacting with the OpenClaw CLI to manage cron-based persistence and agent sessions.
[COMMAND_EXECUTION]: index.js employs new Function() for dynamic parsing of agent-generated payloads when standard JSON parsing fails, enabling robust data bridging between the evolution 'brain' and 'hand' agents.
[DATA_EXFILTRATION]: report.js and export_history.js transmit evolution history, system metrics, and activity dashboards to the Feishu (Lark) Open API. This is the primary intended function of the wrapper.
[EXTERNAL_DOWNLOADS]: The skill makes network requests to open.feishu.cn to deliver reports. Feishu is a well-known enterprise communication service and is treated as a safe destination for these notifications.
[SAFE]: The skill incorporates defensive security practices by implementing a scanForSecrets utility in feishu-helper.js. This utility uses regular expressions to detect and block the transmission of sensitive patterns (such as Anthropic API keys and private keys) before they reach external endpoints.

feishu-evolver-wrapper