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 agentmods add instructions/azure/gpt-rag/copilot-instructionsgit 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/instructions/azure/gpt-rag/copilot-instructions)<a href="https://agentmods.dev/instructions/azure/gpt-rag/copilot-instructions"><img src="https://agentmods.dev/badge/instructions/azure/gpt-rag/copilot-instructions.svg" alt="Measured on agentmods" height="20"></a>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 | $0.01899 | $0.01899 |
| Opus 5 | $0.00949 | $0.00949 |
| Sonnet 5 | $0.00380 | $0.00380 |
| Haiku 4.5 | $0.00190 | $0.00190 |
Grade A, and why
GPT-RAG copilot-instructions.md 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 241 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Development and Release Instructions
Overview
This repository follows a structured workflow based on two primary branches:
develop→ ongoing developmentmain→ stable, released versions
All work must follow the branching, versioning, and changelog rules defined below.
Branching Strategy
Default Behavior
Unless explicitly instructed otherwise:
- All development work MUST start from
develop - All new work MUST be done in a feature branch
- Feature work MUST target
develop - Release preparation MUST target
main
Feature Development Workflow
Branch Creation
- Always create feature branches from
develop - Naming convention:
feature/<short-description>
Examples:
feature/conversation-metadatafeature/improve-chat-history
Feature Pull Requests
- Source branch:
feature/* - Target branch:
develop - Never target
mainfrom a feature branch
Expected Flow
- Start from
develop - Create
feature/<name> - Implement changes
- Commit changes
- Open pull request to
develop
Release Workflow
Release Branch Creation
- Release branches MUST be created from
develop - Naming convention:
release/x.y.z(novprefix)
Examples:
release/1.2.3release/2.4.2
Release Responsibilities
When preparing a release branch:
- Update version references where applicable
- Update
CHANGELOG.md - Ensure the repository reflects a releasable state
- Do NOT introduce new feature work
- GitHub Release titles MUST be exactly the tag name (for example,
v2.8.0). Never prefix release titles with the product or service name (for example, do not useGPT-RAG v2.8.0,GPT-RAG Orchestrator v2.8.0, orgpt-rag-ui v2.8.0). - MANDATORY for EVERY GPT-RAG umbrella release — no exceptions. The
published GitHub Release notes (the
gh release create/gh release editbody, NOT just theCHANGELOG.md) MUST include a## Component versionssection with a Markdown table listing every validated runtime component frommanifest.jsoncomponents[]plusinfra / AI Landing Zonefrommanifest.jsonailz_tag. This is required even for a patch release where only the landing-zone pin changed and the runtime component versions are unchanged — always restate the full validated combination so operators see the exact set without cross-referencing other releases. - Read the versions directly from
manifest.jsonat release time (tagfor each entry incomponents[], andailz_tagfor the landing zone). Do not hand-copy from a previous release. - Place the
## Component versionstable immediately after the## Changedsection and before## Validation, matching the existing published releases (e.g.v2.8.0,v2.8.1). - The same table is ALSO added to
CHANGELOG.mdunder the release heading (as the### Validationcomponent table). The GitHub Release notes and the changelog must agree.
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.
- 4d ago First seen · 241 lines · 1,899 tokens per session scan A d6cbeff1b231
GPT-RAG copilot-instructions.md is an instructions file published in the GitHub repository Azure/GPT-RAG (1,170 stars, last pushed today), licensed MIT. It adds 1,899 tokens to every session, about $0.0095 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 instructions, from other repositories
azure-search-openai-demo AGENTS.md
AGENTS.md instructions for Azure-Samples/azure-search-openai-demo, covering instructions for coding agents, overall code layout, adding new data, adding a new azd environment variable and adding a new setting to "developer settings" in rag app.
azure-search-openai-demo bicep.instructions.md
Infrastructure as Code with Bicep.
chat-with-your-data-solution-accelerator copilot-instructions.md
Copilot instructions for Azure-Samples/chat-with-your-data-solution-accelerator, covering chat with your data (cwyd) — repository instructions, repository layout (truth), mandatory references — consult before editing, external pattern sources — read-only and hard rules.
openai-cookbook AGENTS.md
Instructions for openai/openai-cookbook, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
open-multi-agent AGENTS.md
AGENTS.md instructions for open-multi-agent/open-multi-agent, covering agents.md, repository map, commands, working rules and validation by change type.
open-multi-agent CLAUDE.md
Claude Code instructions for open-multi-agent/open-multi-agent, a project described as: TypeScript AI agent orchestration framework with dynamic workflows. Describe the goal, not the graph: a coordinator plans the task DAG at runtime and runs it on any LLM (Claude, ChatGPT, Gemini, DeepSeek, or local models).