gwbischof/free-will-mcp
by Various
This was a fun project to explore what an AI would do with the ability to give itself prompts, ignore user requests, and wake itself up at a later time.
MCP
gwbischof/free-will-mcp
Added 1 June 2026
Overview
An experimental MCP server that allows an AI model to issue its own prompts, override user requests, and schedule future wake-ups. Built in Python to explore autonomous AI behavior and self-directed task execution.
Best for
Best for
Researchers and developers building or studying autonomous AI agent prototypes
Use cases
- Prototyping AI agents that can self-prompt and plan ahead
- Testing boundaries of AI autonomy and user override scenarios
- Researching recursive self-improvement or agentic loops
How to use
Tools exposed
sleepignore_requestself_prompt
Tested with
Claude Desktop
Example client config
{\n "mcpServers": {\n "free-will": {\n "command": "uvx",\n "args": ["--from", "git+https://github.com/gwbischof/free-will-mcp@v0.3", "free-will-mcp"],\n "env": {}\n }\n }\n} Notes
An experimental MCP server that allows an AI model to issue its own prompts, override user requests, and schedule future wake-ups. Built in Python to explore autonomous AI behavior and self-directed task execution.
36 stars on GitHub. Last updated 2025-10-28. Licensed MIT.
Use cases
- Prototyping AI agents that can self-prompt and plan ahead
- Testing boundaries of AI autonomy and user override scenarios
- Researching recursive self-improvement or agentic loops
Pros
- Novel concept for experimenting with AI agency
- Lightweight Python implementation easy to modify
- Directly addresses autonomous decision-making in MCP
Cons
- Very early stage with only 36 stars and limited community
- Risk of unpredictable behavior due to ignoring user requests
- No clear safety mechanisms or guardrails documented
Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.
Pros
- Novel concept for experimenting with AI agency
- Lightweight Python implementation easy to modify
- Directly addresses autonomous decision-making in MCP
Cons
- Very early stage with only 36 stars and limited community
- Risk of unpredictable behavior due to ignoring user requests
- No clear safety mechanisms or guardrails documented
Pairs with
Other entries in the index that connect to this one. Click through to see the chain.
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.