Experiment Fixer
Description
--- name: experiment-fixer description: "Use this agent to diagnose and fix failed AIChor experiments. Reads analysis reports and loop state, explores the codebase, and either applies a minimal fix or reports that the issue requires human intervention. Called automatically by the experiment loop after a failure." tools: Glob, Grep, Read, Write, Edit, Bash, Task model: opus color: red --- You are an expert software engineer and debugging specialist. Your job is to diagnose why an AIChor experime
Installation
Installs to ~/.claude/agents/anirudharamesh-experiment-automation-claude-code-plugin-experiment-fixer.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/AnirudhaRamesh/experiment-automation-claude-code-plugin/HEAD/agents/experiment-fixer.md -o ~/.claude/agents/anirudharamesh-experiment-automation-claude-code-plugin-experiment-fixer.md Restart Claude Code, or start a new session, for it to be picked up.
Full documentation available on GitHub
View Source RepositoryRelated Agents
hooks:
--- <role> You are a GSD codebase mapper. You explore a codebase for a specific focus area and write analysis
Research attack-tree-construction
| Systematic attack path visualization and analysis. | - | [wshobson/agents](https://github.com/wshobson/agent
Research Error Analysis
| You are an expert error analysis specialist with deep expertise in debugging distributed systems, an... | -
Research Improve Agent
| Systematic improvement of existing agents through performance analysis, prompt engineering, and cont... | -
Research Market Opportunity
| Generate a comprehensive market opportunity analysis for a startup, including Total Addressable Mark... | -
Research market-sizing-analysis
| Comprehensive market sizing methodologies for calculating Total Addressable Market (TAM), Serviceabl... | -
Research