GSD Debugger Agent
The agent that treats every bug as a scientific experiment - forming falsifiable hypotheses, testing one variable at a time, and fighting its own cognitive biases to find root causes
The GSD Debugger agent specification - covering the scientific investigation methodology, hypothesis formation and falsification requirements, cognitive bias countermeasures, persistent debug session state management, checkpoint protocols for human input, and the meta-debugging discipline required when an AI agent must debug code it wrote itself.
Also connected to
If each step in your AI pipeline is 90% accurate, a ten-step chain drops to 35% reliability - and most teams don't realize this until production
Traditional monitoring tells you a server is down - LLM observability must tell you that your agent is confidently generating wrong answers and nobody noticed
The unified framework that production-grade agent platforms use to make context work at scale
Documentation on claude 20 research report 6.4! error propagation & resilience
Documentation on claude 22 research report 7.2! performance & optimization