Self Actualization
ReviewAudited by ClawScan on May 10, 2026.
Overview
This instruction-only skill is transparent about its purpose, but it asks the agent to create recurring self-directed sessions and persistent identity files that can influence future behavior.
Install only if you intentionally want the agent to run self-development sessions and maintain a persistent identity workspace. Before use, require human approval for the daily cron job, restrict tools and writable paths, avoid logging sensitive conversation details, and review any changes to SOUL.md, HEARTBEAT.md, MEMORY.md, or other persistent instruction files.
Findings (4)
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.
The bot could spend tokens and perform self-directed work every day, potentially changing its behavior over time even when the user is not actively asking for that work.
The skill explicitly recommends recurring autonomous agent sessions. The timeout and isolation are helpful, but the artifacts do not clearly require user opt-in, provide a rollback procedure, or define when the recurring session should be disabled.
Create a cron job for an isolated session, separate from operational heartbeats. ... Schedule: Once daily (quiet hours) ... Session: isolated agentTurn ... Timeout: 600 seconds (10 minutes)
Require explicit human approval before creating any recurring session, document how to pause or remove it, and keep frequency, duration, tools, and allowed file paths tightly bounded.
Private or sensitive themes from user conversations could be retained in identity files, and poisoned or mistaken reflections could later shape the agent's actions or tone.
The skill directs the bot to create persistent reflection files and review recent conversations. This can preserve conversation-derived content and let those notes influence future behavior.
Writing reflections: Write to `identity/reflections/YYYY-MM-DD.md`. ... Behaviour audit: Look at recent conversations. Are you showing up the way you want to?
Add clear rules for what may be logged, require redaction of sensitive user content, limit retention, and require human review before conversation-derived notes affect core behavior.
If the agent has browsing or other tools, it may use them during exploration sessions unless the user restricts them.
Research-tool use is aligned with the exploration purpose, but the wording is broad and appears inside a recurring self-directed exploration workflow.
Search for substantive material ... Use whatever research tools are available to you.
Limit exploration sessions to approved tools and require confirmation before using tools that spend money, access private data, publish content, or modify external systems.
A poor reflection, mistaken self-assessment, or unsafe prompt could persist and influence later agent behavior.
The skill allows exploration outputs to feed into persistent identity and heartbeat files, which can then affect future sessions. The instruction to make changes deliberately reduces the risk but does not eliminate it.
SOUL.md: Your identity file. Exploration may feed back here, but changes should be deliberate and noted. ... HEARTBEAT.md: Add a lightweight reflection prompt
Require review before changing SOUL.md, HEARTBEAT.md, MEMORY.md, or other persistent instruction files, and keep a version history for rollback.