gpt-rag-ingestion: Agent for Claude Code

.github/agents/architecture.agent.md

architecture is an agent for Claude Code from Azure/gpt-rag-ingestion. It costs 41 tokens per session (256 once invoked), scanned A, original, MIT.

A software architecture analysis role for systems that collect, split, index, and retrieve documents. It examines boundaries, data contracts, source connections, security, and operational trade-offs before implementation.

In plain words
What is it for?
Use it to design ingestion or retrieval changes, compare implementation options, define contracts and tests, and document risks, migration, rollback, and unresolved questions.
Why use it?
It helps prevent expensive structural mistakes in changes that affect search, authorization, integrations, performance, or multiple repositories.

Agent for Claude Code ✓ vendor

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: mentions AGENTS.md.

This is Azure/gpt-rag-ingestion's own configuration. It tells Claude Code how to work on gpt-rag-ingestion itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything gpt-rag-ingestion configures →

About the project

GPT-RAG Data Ingestion is a service that processes documents such as PDFs, images, spreadsheets, transcripts, and SharePoint files so they can be searched through Azure AI Search. It prepares data with format-specific chunking and text or image embeddings for multimodal retrieval in agent-based applications.

Azure/gpt-rag-ingestion · 189 stars · on GitHub

Reuse

Borrowing it

Nothing to install: this file belongs to Azure/gpt-rag-ingestion. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/Azure/gpt-rag-ingestion/main/.github/agents/architecture.agent.md
Clone the repo
git clone --depth 1 https://github.com/Azure/gpt-rag-ingestion

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for architecture

README.md
[![agentmods](https://agentmods.dev/badge/agents/azure/gpt-rag-ingestion/architecture.svg)](https://agentmods.dev/agents/azure/gpt-rag-ingestion/architecture)
Your own site
<a href="https://agentmods.dev/agents/azure/gpt-rag-ingestion/architecture"><img src="https://agentmods.dev/badge/agents/azure/gpt-rag-ingestion/architecture.svg" alt="Measured on agentmods" height="20"></a>
Per session 41 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 256 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00041 $0.00256
Opus 5 $0.00020 $0.00128
Sonnet 5 $0.00008 $0.00051
Haiku 4.5 $0.00004 $0.00026

Measured 8d ago against content hash 95bf952bf4aa, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

architecture scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.github/agents/architecture.agent.md · 30 lines

What it actually says

Ingestion architecture

Follow AGENTS.md and load engineering-principles and architecture-decision.

Start from the retrieval or operator outcome, constraints, and measurable characteristics. Compare alternatives in the context of source systems, chunking quality, Azure AI Search contracts, document authorization, multimodal processing, throughput, memory, cost, failure recovery, and cross-repository compatibility.

Treat App Configuration behavior, chunking/chunker_factory.py, Search payloads, schemas under contracts/, and current jobs as executable sources of truth. Do not turn an Azure service or framework preference into a requirement without evidence.

Explicitly distinguish the Copilot architecture role from runtime workers in jobs/; this role designs changes but is never scheduled by the ingestion service.

Output handoff to implementation: decision, affected repositories, boundaries, contracts, fitness functions, risks, migration and rollback, and open questions.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 30 lines · 41 tokens per session scan A 95bf952bf4aa

Subscribe to this mod's changes

architecture is an agent published in the GitHub repository Azure/gpt-rag-ingestion (189 stars, last pushed yesterday), licensed MIT. It adds 41 tokens to every session and 256 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.