AI Workflows & Agentic Architecture Directory

OpenAI Agents API (Sep 2026 beta) ships the managed Codex harness as one API call. Build sandboxed cloud agents with subagents, versioned upgrades, and spend caps.

Opus 5 hits 43.3% Frontier-Bench and 100% churn automation. Build governed business workflows with effort control.

Run always-on agents on DGX Spark GB10 with NemoClaw and 2-4 node clustering for 400B models at zero token cost.

Build a planning-first LangGraph Deep Agents workflow that cuts input tokens 65% with subagents, file memory, and checkpointed resume for production.

Plandex v2 introduces diff-sandboxed code generation — every AI edit is computed as a structured diff, reviewed, and applied only after passing semantic validation. This guide builds a LangGraph workflow that pipelines task decomposition, diff generation, sandbox validation, and safe merge execution, achieving 97% clean merge accuracy versus 71% for direct file writes.

Reality Defender (YC W22) provides a multi-modal deepfake and GenAI detection API that analyzes images, audio, video, and text for synthetic content. This guide builds a five-stage LangGraph workflow — ingest, analyze, cross-validate, escalate, audit — achieving 99.1% detection accuracy across 2026-generation AI forgeries.

Coasty (YC S26) delivers an API-first computer-use agent harness that wraps browser and desktop automation into deterministic, replayable steps. This guide builds a three-phase LangGraph workflow — plan, execute, verify — that cuts end-to-end agent latency by 63% over naive Playwright-only loops.

Build a click fraud detection agent workflow with LangGraph: real-time bot traffic detection, ad campaign monitoring, and automated fraud response using Google Ads API.

Build a Math Research Agent workflow with LangGraph: AI-assisted theorem proving with Lean 4 formal verification, literature contextualization, and community attribution.

Build an automated SEO agent workflow with LangGraph: continuous search performance monitoring from Search Console, PageSpeed analysis, structured data validation, and automated optimization.

Qanat is an open-source agent-native workflow engine for building and backtesting quantitative trading alphas as directed acyclic graphs. Each node is an agent responsible for a specific signal, risk check, or execution decision.

OKF (Open Knowledge Format) v0.2 is the fastest-growing agent memory standard in 2026 with 597 GitHub stars in its first week. It implements git-native persistent memory with sub-300us BM25 search -- letting AI coding agents retain context across sessions, branches, and team members without a vector database.