AI Workflow Reference Hub
Find the site’s stable reference pages for AI platform mapping, workflow placement, prompt and instruction layers, verification methods, terminology, diagrams, engineering protocols, and content navigation.
Find the right reference page
AI workflow placement by platform
Start here when a guide says “configure this workflow.” Before choosing a platform page, separate stable instructions, source material, reusable workflows, runtime prompts, and verification gates so each part is placed on the correct AI-tool surface.
Platform mapping
ChatGPT workflow placement
Projects, project instructions, files/sources, GPTs, Skills, Custom Instructions, Memory, and runtime prompts.
Platform mapping
Claude workflow placement
Projects, project instructions, Project Knowledge, Skills, Claude Code surfaces, memory/personalization boundaries, and runtime chat input.
Platform mapping
Gemini workflow placement
Gems, Gem instructions, Knowledge files, Gemini instructions, AI Studio, Gemini API, Files API, File Search, and Vertex AI surfaces.
Platform mapping
API / internal systems workflow placement
System/developer instructions, retrieval, file ingestion, tools, orchestration, approvals, validation, evals, logging, and application state.
Core reference areas
Open the reference surface that matches the kind of lookup material you need.
Canonical layers
Prompt layers & policy mapping
Definitions for policies, system/developer instructions, user runners, components, workflows, stacks, and API/agent layer boundaries.
Verification methods
Hallucination mitigation methods
Reference index for verification methods used to reduce unsupported claims and check model outputs against evidence.
System models
Diagrams
Architecture, request assembly, trust boundary, connector, tool-routing, and attack-surface diagrams.
Site navigation
Content map
Stable map of public site surfaces and the shortest paths by task, content type, and account need.
Terminology
Orders of intentionality
Reference definition for recursive mindreading and Theory-of-Mind terminology used across the site.
Use Reference from a workflow
These are the common lookup questions that should point to Reference instead of repeating the same platform or terminology explanation inside every guide.
Where should I put this in ChatGPT?
Use the ChatGPT placement map for Projects, GPTs, Skills, files/sources, Memory, and runtime prompts.
Where should I put this in Claude?
Use the Claude placement map for Projects, Project Knowledge, Project Instructions, Skills, and runtime chat input.
Where should I put this in Gemini?
Use the Gemini placement map for Gems, Knowledge files, Gemini instructions, AI Studio, Gemini API, and Vertex AI surfaces.
Where should this live in an API or internal agent system?
Use the API placement map for system/developer instructions, retrieval, tools, orchestration, approvals, validation, evals, logs, and application state.
Which workflow layer is this?
Use prompt-layer mapping to separate policies, instructions, runners, components, workflows, stacks, and API/agent boundaries.
Which verification method applies?
Use the verification methods index when the workflow needs claim checking, evidence review, or hallucination mitigation.
Which system model explains this?
Use diagrams for request assembly, trust boundaries, connectors, tool routing, provenance, and attack-surface models.