jasontang-ai

Next-generation autonomous development orchestrator with cognitive feedback loops. Executes complex multi-step features from PRDs through iterative agent sessions with quality verification, context synthesis, and recursive learning. Use when implementing features that require multiple stories, exceed single context windows, or need autonomous execution with quality guarantees. Replaces manual iteration with intelligent orchestration.

Generate structured Product Requirements Documents through guided questions. Creates well-formed PRDs with user stories, acceptance criteria, and technical approach. Use when planning new features, creating specifications, or when user asks to create/generate a PRD.

Convert markdown PRD to prd.json format for Ralph Zero autonomous execution. Validates story structure, checks dependencies, ensures right-sizing, and generates validated JSON. Use when you have a PRD markdown file and need to prepare it for autonomous development.

Use this to mathematically guarantee zero race conditions and logical correctness via Dafny formal verification.

SlopBench soft oracle derived from Andrej Karpathy's behavioral guidelines for LLM coding (https://github.com/forrestchang/andrej-karpathy-skills). Enforces simplicity-first design, surgical changes, and goal-driven execution as structural constraints. Pipeline: https://github.com/davidkimai/specoracle

SlopBench soft oracle enforcing the Zen of Python (PEP 20) as structural quality constraints during LLM code generation. Conditions outputs toward simplicity, flatness, explicitness, and local auditability. Pipeline: https://github.com/davidkimai/specoracle