fable-harness
NewMake Claude Code work like a disciplined engineer: OODA, multi-party adversarial review, model routing, fail-then-pass. Distilled from Fable to reinforce the Opus harness.
Summary
This skill enforces a disciplined engineering workflow in Claude Code, incorporating OODA loops, multi-party adversarial review, model routing, and a fail-then-pass strategy.
- It helps developers produce robust, well-reviewed code by simulating rigorous peer review and iterative improvement.
Install & Usage
mkdir -p .claude/skillsAdd the configuration to .claude/skills/fable-harness.md
/fable-harnessUse Cases
Usage Examples
/fable-harness Review the changes in this PR for security vulnerabilities and design flaws using adversarial review.
/fable-harness Refactor the function calculate_total to handle edge cases, using OODA loops until all tests pass.
/fable-harness Debug the failing test in test_user_auth.py by applying fail-then-pass and multi-party review.
Security Audits
Frequently Asked Questions
What is fable-harness?
This skill enforces a disciplined engineering workflow in Claude Code, incorporating OODA loops, multi-party adversarial review, model routing, and a fail-then-pass strategy. It helps developers produce robust, well-reviewed code by simulating rigorous peer review and iterative improvement.
How to install fable-harness?
To install fable-harness: create the skills directory (mkdir -p .claude/skills), then add the config to .claude/skills/fable-harness.md. Finally, /fable-harness in Claude Code.
What is fable-harness best for?
fable-harness is a other categorized under General. It is designed for: code-review. Created by Miguok.
What can I use fable-harness for?
fable-harness is useful for: Review a pull request with adversarial critique from multiple simulated personas.; Refactor a complex function by iterating through OODA loops until all failure modes are addressed.; Generate code with automatic model routing to select the best AI model for each subtask.; Debug a failing test by applying fail-then-pass methodology to identify root causes.; Design a system architecture with structured adversarial review to uncover hidden assumptions.; Optimize performance bottlenecks using iterative OODA cycles and multi-perspective analysis..