Diagrams

This index lists technical diagrams.

Conventions (page-level contract):

• Each diagram page should include a caption and short alt text.
• Complex diagrams should additionally include a text alternative (long description) to meet the intent of WCAG text-alternatives guidance.

Request assembly & context selection

• Request assembly / context selection — attack surface (summary)
  Summary schematic of the request-assembly boundary highlighting common failure modes: instruction injection, precedence confusion, truncation, provenance loss, and memory poisoning.
• Effective-privileges invariant — provenance & replay failure modes
  Reference diagram showing two common failures: chat-derived context influencing effective privileges, and model output persistence/replay into later context; highlights the server-side AuthZ + authoritative state invariant.
• ChatGPT request assembly architecture
  Request assembly/context selection schematic with annotated risk points across tools, memory, retrieval/caching, and observability.
• ChatGPT request assembly architecture (detailed)
  Detailed author-mapped diagram for request assembly, context inputs hub, tool loops, optional persistence loops, observability, and R1–R8 checkpoints.
• Provenance Boundary Failure — Prompt Assembly Diagram
  Reference model of provenance boundary failure in prompt assembly: authority confusion and cross-turn drift.

Back to categories

Connectors & tool routing

• Web tool access — prompt injection risk map
  Reference schematic of prompt-injection risk in web tool access: routing manipulation and downstream failure modes (content-based instructions, source poisoning, query leakage, unbounded consumption, improper output handling).
• ChatGPT connectors: basic vs agent mode
  Schematic comparing basic-mode hinting vs connector/agent-mode routing to external services, with control/risk callouts.

Back to categories