GPT-RAG is an Azure solution accelerator for building agentic retrieval-augmented generation systems, where agents use connected organizational sources to ground their answers. It is for organizations deploying secure RAG applications with Microsoft Foundry, Microsoft Agent Framework, and Azure infrastructure. The catalogue add-ons provide agents, skills, and instructions for deploying and working with the accelerator.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add Azure/GPT-RAG --skill architecture-decisiongit clone --depth 1 https://github.com/Azure/GPT-RAGWrote 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.
[](https://agentmods.dev/skills/azure/gpt-rag/architecture-decision)<a href="https://agentmods.dev/skills/azure/gpt-rag/architecture-decision"><img src="https://agentmods.dev/badge/skills/azure/gpt-rag/architecture-decision/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/azure/gpt-rag/architecture-decision"><img src="https://agentmods.dev/badge/skills/azure/gpt-rag/architecture-decision.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00042 | $0.00209 |
| Opus 5 | $0.00021 | $0.00105 |
| Sonnet 5 | $0.00008 | $0.00042 |
| Haiku 4.5 | $0.00004 | $0.00021 |
Grade A, and why
architecture-decision 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 9d 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.
What it actually says
GPT-RAG architectural decision
- Load the relevant
engineering-principlesreferences. - Define context, constraints, affected repositories, and up to five prioritized characteristics with measures.
- Compare at least two viable alternatives and the option of not changing.
- Evaluate identity, authorization, network isolation, component compatibility, cost, operation, migration, and reversibility.
- Record the decision under
docs/adr/using the ADR template. - Define fitness functions, adoption order, rollback or roll-forward, and a review trigger.
Do not turn a tool or framework preference into an architectural requirement. When evidence is missing, record a time-bounded investigation and its decision criterion instead of guessing.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 22 lines · 42 tokens per session scan A 338048871929
architecture-decision is a skill published in the GitHub repository Azure/GPT-RAG (1,173 stars, last pushed today), licensed MIT. It adds 42 tokens to every session and 209 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.
Other skills, from other repositories
bootstrap-realtime-eval
Bootstrap a new realtime eval folder inside this cookbook repo by choosing the right harness from examples/evals/realtimeevals, scaffolding prompt/tools/data files, generating a useful README, and validating it with smoke, full eval, and test runs. Use when a user wants to start a new crawl, walk, or run realtime eval…
verifier
Double-model verification gate before declaring a build/research/refactor "done". Runs a fresh Anthropic (Claude) review AND an independent OpenAI/Azure critic against the stated goal + criteria, then reconciles. Use before announcing completion of medium/major builds, refactors, major fixes, research deliverables…
Agent Developer
Microsoft Agent Framework + Foundry Toolbox MCP integration specialist.
tmux
Remote-control tmux sessions for interactive CLIs by sending keystrokes and scraping pane output.
summarize
Summarize or extract text/transcripts from URLs, podcasts, and local files (great fallback for “transcribe this YouTube/video”).
foundry-config-setup
Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.