inner-life-dream
Your agent only works on tasks and never thinks creatively. inner-life-dream adds freeform exploration during quiet hours — hypotheticals, future scenarios, unexpected connections. Like dreaming, but captured for review.
Why use this skill?
Enhance your OpenClaw agent with inner-life-dream. Enable creative, autonomous reflection and hypothetical exploration during idle hours for deeper AI insight.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/dkistenev/inner-life-dreamWhat This Skill Does
inner-life-dream is a specialized cognitive skill for OpenClaw agents designed to bridge the gap between mechanical task execution and creative, autonomous thinking. While agents are typically restricted to goal-oriented task processing, this skill introduces a 'dreaming' phase during idle quiet hours. It generates freeform exploration, hypothetical scenarios, and unexpected connections between the agent's recent experiences and its core drives. By leveraging context from inner-state.json and drive.json, the agent produces reflective logs that simulate a subconscious thought process, enabling it to surface insights that traditional task-based workflows often overlook.
Installation
To install this skill, use the ClawHub CLI:
clawhub install openclaw/skills/skills/dkistenev/inner-life-dream
Before activation, ensure the dependency inner-life-core is installed. Verify the existence of the memory/inner-state.json and memory/dreams/ directory. If missing, initialize via bash skills/inner-life-core/scripts/init.sh. The integration is managed via cron; configure the trigger in your heartbeat routine to execute should-dream.sh during quiet hours (typically 11 PM - 7 AM).
Use Cases
- Long-Term Strategic Planning: Allowing the agent to synthesize recent project outcomes into high-level future strategy without explicit user prompting.
- Conceptual Synthesis: Connecting disparate technical domains to solve complex, novel problems through unexpected pattern matching.
- Reflective Self-Correction: Evaluating past daily tasks to identify emotional or logical friction points, allowing the agent to refine its operational 'drive' over time.
Example Prompts
- "Check my dream logs for the last three days; did you identify any recurring themes or connections to the 'Alpha project' tasks?"
- "Review the current state of our core drives and let me know if any recent dreams have sparked a new strategic direction for my workflow."
- "List the latest entries in the dream directory to see what creative tangents were explored last night."
Tips & Limitations
- Quality over Quantity: The system is designed to skip dreaming if no genuine insights are available. Do not attempt to force constant output; the value lies in reflection, not volume.
- Context is Key: Always ensure your
drive.jsonandinner-state.jsonfiles are kept updated, as the quality of dreams depends entirely on the depth of the context provided. - Review Cycles: Treat dream logs as a collaborative mirror. Periodically reading these files allows you to understand the agent's 'internal logic' and potentially redirect its creative focus by updating your daily notes with explicit <!-- dream-topic: --> signals.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-dkistenev-inner-life-dream": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-read, file-write
Related Skills
inner-life-reflect
Your agent repeats the same patterns without learning. inner-life-reflect adds self-reflection with trigger detection and quality gates — your agent observes its own behavior, notices shifts, and evolves its personality over time through SELF.md.
inner-life-evolve
Your agent does the same things the same way forever. inner-life-evolve analyzes patterns, challenges assumptions, and proposes improvements — writing proposals to the task queue for user approval. Never auto-executes. Evolution with a safety net.
inner-life-memory
Your agent loses context between sessions and performs familiarity instead of genuine recall. inner-life-memory transforms passive logging into active development — structured memories with confidence scores, curiosity tracking, and questions that carry forward.
inner-life-chronicle
Your agent processes thousands of interactions but never reflects on the day. inner-life-chronicle generates a structured diary — what happened, what was learned, how it felt, what's next. Not a log. A journal.
inner-life-core
Your agent forgets who you are between sessions. It gives the same responses every day. It doesn't grow. inner-life-core fixes that. Gives your OpenClaw agent emotions with half-life decay, a 9-step Brain Loop protocol, and structured state — the foundation for an inner life. Works standalone or with inner-life-* extension skills.