auto-daily-summary
ReviewAudited by ClawScan on May 10, 2026.
Overview
The skill largely does what it claims, but it creates persistent daily jobs for every agent and builds shell commands unsafely, so it should be reviewed before use.
Run this only if you want every discovered OpenClaw agent to receive a daily summary prompt indefinitely. Review the script first, consider limiting it to selected agents, and know how to list and delete the created cron jobs.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A single run can leave daily autonomous prompts active for all agents until the user manually finds and removes those cron jobs.
The script iterates over every discovered agent and adds a recurring OpenClaw cron job. The artifacts do not show per-agent opt-in, a confirmation step, an expiration, or a cleanup command.
for agent in agents: ... create_cron_job(agent_id, agent['workspace'], system_tz) ... f'--cron "30 23 * * *" ' ... f'--agent {agent_id} 'Add a dry-run or confirmation step, support selecting specific agents, and document how to remove the created jobs. Users should inspect `openclaw cron list` after running it.
If an agent id, workspace path, or timezone value contains shell metacharacters, running the setup script could execute unintended local commands.
The script uses shell=True and interpolates values derived from local agent data, workspace paths, and timezone settings into a shell command without robust quoting or validation.
result = subprocess.run(command, shell=True, ...) ... f'--tz "{timezone}" ' ... f'--agent {agent_id} ' ... f'--message "{message}" 'Use `subprocess.run([...], shell=False)` with an argument list, and validate or safely escape agent ids, workspace paths, and timezone values.
Task details, collaboration notes, and possibly sensitive work information may be retained in agent memory files; inaccurate summaries could also persist.
The skill intentionally causes agents to write recurring summaries into persistent workspace diary files, which may later be reused as memory/context.
write them to their diary file [workspace]/memory/daily/YYYY-MM-DD.md... cross-agent collaborations, and learning insights
Confirm that this diary location and retention behavior are acceptable, and periodically review or prune generated memory files.