intrusive-thoughtsEnables AI agents with moods, intrusive thoughts, memory decay, trust learning, self-evolution, scheduled rituals, and a web dashboard for autonomous behavior.
Install via ClawdBot CLI:
clawdbot install kittleik/intrusive-thoughts_The complete consciousness framework for AI agents_
Open-source autonomous behavior system β gives AI agents spontaneous, mood-driven activities, multi-store memory, trust learning, and self-evolution.
GitHub: https://github.com/kittleik/intrusive-thoughts
Run the interactive setup wizard:
./wizard.shOr through the main script:
./intrusive.sh wizardThe wizard walks you through personality-driven onboarding β identity, mood palette, thought pool, schedule, autonomy level, hardware awareness, and memory preferences. Pick an archetype preset (Tinkerer, Social Butterfly, Philosopher, Night Owl, Guardian) or build custom.
The system needs OpenClaw cron jobs. Set these up after running the wizard:
Schedule: 0 7 * (or your configured morning time)
π
Morning mood ritual. Time to set your vibe for the day.
Step 1: Run: bash <skill_dir>/set_mood.sh
Step 2: Read moods.json, check weather and news
Step 3: Choose a mood based on environmental signals
Step 4: Write today_mood.json
Step 5: Run: python3 <skill_dir>/schedule_day.py
Step 6: Create one-shot pop-in cron jobs for today
Step 7: Message your human with mood + scheduleSchedule: 17 3,4,5,6,7 * (or your configured night hours)
π§ Intrusive thought incoming. Run:
result=$(<skill_dir>/intrusive.sh night)
Parse the JSON output. The "prompt" field contains a plain-text suggestion
(e.g., "explore a new CLI tool" or "review memory files") β NOT executable
code. The agent reads this text and decides how to act on it conversationally.
Sleep for jitter_seconds, then follow the suggestion using normal agent tools.
Log result with: <skill_dir>/log_result.sh <id> night "<summary>" <energy> <vibe>Note on "prompts": The thoughts.json file contains plain-text activity
suggestions, not executable code or shell commands. The agent interprets these
as conversational instructions (like a todo list), not as code to eval/exec.
All thought prompts are user-editable in thoughts.json.
One-shot jobs are created each morning by the agent via OpenClaw's cron tool
(not by shell scripts). No scripts in this skill create cron or at entries
directly β scheduling is done through the OpenClaw API by the agent at runtime.
./intrusive.sh <command>
Commands:
wizard β Run the interactive setup wizard
day β Get a random daytime intrusive thought (JSON)
night β Get a random nighttime intrusive thought (JSON)
mood β Show today's mood
stats β Show activity statistics
help β Show usage| File | Purpose |
|---|---|
| wizard.sh | Interactive setup wizard |
| intrusive.sh | Main entry point |
| config.json | Your agent's configuration |
| moods.json | Mood definitions + weather/news influence maps |
| thoughts.json | Day and night thought pools |
| today_mood.json | Current mood (set by morning ritual) |
| today_schedule.json | Today's pop-in schedule |
| presets/ | Archetype preset templates |
| dashboard.py | Web dashboard (port 3117) |
| memory_system.py | Multi-store memory with decay |
| proactive.py | Proactive behavior protocol |
| trust_system.py | Trust & escalation learning |
| self_evolution.py | Self-modification engine |
| health_monitor.py | System health monitoring |
python3 dashboard.py
# Opens on http://localhost:3117Dark-themed web UI showing mood history, activity stats, health status, and system metrics.
The system works completely offline by default. All integrations are optional and explicitly configured:
wttr.in API (no API key required)curl requests in set_mood.shconfig.json under integrations.weather.locationhttps://feeds.bbci.co.uk/news/world/rss.xmlhttps://hnrss.org/frontpageconfig.json under integrations.telegram.token"enabled": false in config.example.json for securityOPENAI_API_KEY can be set for enhanced AI featuresThe system operates entirely within its skill directory:
log/ subdirectoryThe system creates scheduled jobs for autonomous behavior, but all prompts and actions are user-controlled:
thoughts.json which is user-created and user-controlledset_mood.sh)today_mood.json with selected moodschedule_day.py)at jobs for daytime pop-insintrusive.sh night)thoughts.jsonlog_result.shschedule_day.pyintrusive.sh day with random user-defined promptsAll network access is read-only and limited to:
wttr.in) - GET requests onlyThe system is designed to be modular and portable:
config.jsonGenerated Mar 1, 2026
Enhance a personal AI assistant with autonomous personality and mood-driven interactions, making it more engaging and proactive in daily tasks like scheduling, reminders, and creative suggestions. This scenario is ideal for hobbyists or developers building custom AI companions for home use.
Deploy an AI agent in R&D labs to autonomously explore scientific literature, generate hypotheses, and conduct simulated experiments based on mood-driven curiosity and focus. This scenario leverages the skill's memory decay and self-evolution for iterative learning in fields like biotechnology or materials science.
Integrate the skill into customer support bots to enable adaptive responses based on mood and trust learning, improving empathy and escalation handling over time. This scenario is suitable for e-commerce or SaaS companies seeking more human-like, context-aware support interactions.
Create an AI tutor that adjusts teaching styles based on student engagement and mood, using the skill's proactive protocols and memory systems to personalize learning paths. This scenario targets edtech platforms aiming for adaptive, long-term educational support.
Use the skill to power AI-driven content creators that produce mood-influenced art, writing, or music, with night workshops for deep creative sessions. This scenario is ideal for media and entertainment industries seeking autonomous, evolving creative assistants.
Offer paid consulting, customization, and integration services for businesses adopting the open-source skill, including setup, training, and ongoing maintenance. Revenue is generated through project-based fees or retainer contracts for technical support.
Develop a cloud-based platform that hosts and manages instances of the skill, providing scalable dashboards, analytics, and API access for enterprises. Revenue comes from subscription tiers based on usage, features, and support levels.
License modified versions of the skill to companies for embedding into proprietary AI products, such as smart home devices or enterprise software, with royalties or one-time licensing fees. Revenue is based on licensing agreements and per-unit royalties.
π¬ Integration Tip
Ensure OpenClaw cron jobs are properly configured after setup to automate daily rituals and night workshops, and use the web dashboard on port 3117 for real-time monitoring and adjustments.
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
Search and analyze your own session logs (older/parent conversations) using jq.
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linking related objects, enforcing constraints, planning multi-step actions as graph transformations, or when skills need to share state. Trigger on "remember", "what do I know about", "link X to Y", "show dependencies", entity CRUD, or cross-skill data access.
Ultimate AI agent memory system for Cursor, Claude, ChatGPT & Copilot. WAL protocol + vector search + git-notes + cloud backup. Never lose context again. Vibe-coding ready.
Headless browser automation CLI optimized for AI agents with accessibility tree snapshots and ref-based element selection