funky-fund-flamingo

What This Skill Does

Funky Fund Flamingo is a specialized self-evolution agent skill designed for OpenClaw users who prioritize measurable growth and financial outcomes. Rather than standard iterative updates, this skill treats your agent's codebase and logic as a profit-generating asset. It functions by systematically inspecting your recent session logs, workspace memory, and existing skill configuration to identify performance bottlenecks, logic failures, or stagnation points. By running targeted 'mutation cycles,' the agent applies repairs to broken flows or expands its own capabilities to better align with your specific objectives. It effectively transforms your agent into a self-optimizing system, ensuring that it isn't just running code, but actively iterating to increase its efficiency and the 'bag'—its ability to fulfill high-value tasks successfully.

Installation

To install this skill, use the ClawHub CLI from your terminal: clawhub install openclaw/skills/skills/icemastert/funky-fund-flamingo Ensure you have the necessary write permissions in your OpenClaw workspace directory, as the skill needs to modify memory files and update local skill logic.

Use Cases

• Automated Debugging: When your agent hits a wall and runtime logs show repeated errors, this skill analyzes the failures and applies fixes to restore profitability.
• Agent Scaling: When your agent is stable but performing repetitive tasks, use it to inject new capabilities that expand the scope of work it can handle.
• Continuous Optimization: Run in --loop mode to ensure the agent is constantly refining its own logic during idle time, effectively polishing its performance in the background.
• Risk-Averse Evolution: Use the --review flag to maintain human oversight, ensuring all significant codebase changes are verified before execution.

Example Prompts

1. "Funky Fund Flamingo, analyze my session logs from this morning. Why are we leaking value on the data extraction tasks? Fix it."
2. "Run a mutation cycle on the codebase, but use --review mode. I want to see your plan to improve my agent's conversion logic before you touch any files."
3. "Enable continuous relay mode. I want the agent to keep evolving its internal strategies while I'm away from the terminal so we can keep making money."

Tips & Limitations

• Safety First: This skill is designed to prioritize stability. If the agent detects systemic instability, it will enter a 'repair-first' state, pausing expansion until the core logic is robust.
• Local Execution: This skill is strictly local. It does not phone home, make HTTP requests, or interact with third-party APIs. All logic is contained within your local repository.
• Resource Management: Avoid running multiple deep mutation cycles simultaneously to prevent file system lock contention. Use --dry-run if you are unsure about the impact of an upcoming cycle.