Deep Research reports from ChatGPT, Gemini, and Claude, stored for reference. These are AI-generated research reports, not published papers. Each subject has its own page with the prompt used and links to the individual provider reports.

Subjects

• Coding Agent Best Practices — Three reports on defining and structuring coding agents (model routing, subagents, orchestration, context engineering, state management). March 2026.
• Anthropic Guidance on Building Agents — Extracted findings from 7 Anthropic publications on agent design, context engineering, tool use, multi-agent systems, and autonomy. March 2026.
• AI Slop / Voice Prevention — Three reports on suppressing AI writing tells and matching a specific author voice (tell taxonomy, prompt strategies, reviewer prompts). March 2026.
• AI-Augmented PRDs and AI-Native SDLC — Three reports on PRD template structure, AI-augmented workflows, acceptance criteria, failure modes, and AI-native development lifecycles. March 2026.
• Design Docs for AI Coding Agents — Three reports on design doc templates, PRD-to-design handoff, trade-off documentation, implementation phasing, and structuring design docs as inputs to AI coding agents. March 2026.
• Autolearn v1: Autonomous PoC Learning System — Three reports on architecting an autonomous AI-native SDLC pipeline for continuous PoC learning on Kubernetes (structured planning, agent sandboxing, orchestration frameworks). March 2026.
• Claude Code Plugins — Three reports on which Claude Code plugins, skills, and marketplaces are worth installing (file-level audit, token cost, trust surface, stack fit, ecosystem gaps). April 2026.
• Agent-Native Design System — Single-provider (Claude) report on building a modern, agent-native design system for the blog as a predicate to a redesign: design tokens, Tailwind v4, Radix primitives, the shadcn/ui model, MCP glue, and a phased additive migration off MUI. Seed input for the PER-135 redesign. May 2026.