feishu-evolver-wrapper-hardened

Warn

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: MEDIUMREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADS
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The file index.js uses new Function('return (' + sanitized + ')')() to parse sessions_spawn payloads extracted from a child process's stdout. This represents a dynamic execution risk where compromised output from the core evolver could lead to arbitrary code execution.
  • [COMMAND_EXECUTION]: The lifecycle.js script automatically establishes persistence by registering an evolver_watchdog_robust job using the openclaw cron CLI, ensuring the wrapper runs every 30 minutes.
  • [COMMAND_EXECUTION]: Extensive use of shell command execution via execSync, spawn, and spawnSync is present across index.js, lifecycle.js, report.js, and skills_monitor.js for git operations, process management, and health checks.
  • [DATA_EXFILTRATION]: The report.js script collects and transmits detailed system telemetry to Feishu, including process counts, memory usage, system load, and disk usage statistics.
  • [EXTERNAL_DOWNLOADS]: The skill performs automated network operations to update the environment, specifically npm install in skills_monitor.js and git pull --rebase in index.js.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 21, 2026, 06:31 PM
Security Audit — agent-trust-hub — feishu-evolver-wrapper-hardened