What is fine-tuning AGENTS.md?

Repository instructions for running fine-tuning work with Microsoft Foundry, a Microsoft service for training or adapting AI models. They explain which guides and examples to read, how to choose managed or interactive training, and when approval is needed before spending money or changing systems.

Use them to select a fine-tuning recipe, prepare its configuration, run managed jobs or interactive sessions, and follow the repository's safety rules.

They reduce setup errors and uncontrolled costs by requiring a preflight review, explicit approval for spending or mutations, and careful handling of secrets and runtime state.

How microsoft-foundry/fine-tuning configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to microsoft-foundry/fine-tuning. 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/microsoft-foundry/fine-tuning/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/microsoft-foundry/fine-tuning

Made for: Codex, OpenCode.

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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/microsoft-foundry/fine-tuning/agents-md/github.svg)](https://agentmods.dev/instructions/microsoft-foundry/fine-tuning/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/microsoft-foundry/fine-tuning/agents-md"><img src="https://agentmods.dev/badge/instructions/microsoft-foundry/fine-tuning/agents-md/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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/microsoft-foundry/fine-tuning/agents-md"><img src="https://agentmods.dev/badge/instructions/microsoft-foundry/fine-tuning/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 1,256 This file is loaded in full into every session.

When invoked 1,256 The same file — it is already loaded in full.

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.

Model	Per session	Once invoked
Fable 5.1	$0.01256	$0.01256
Opus 5.5	$0.00502	$0.00502
Sonnet 5.5	$0.00251	$0.00251
Haiku 4.5	$0.00126	$0.00126

Measured 25d ago against content hash c9b0cec16f74, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

fine-tuning AGENTS.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 25d 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.

AGENTS.md · 90 lines

How it starts

The opening of the file, as written. The whole thing — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Coding-agent execution contract

These instructions apply to the repository. Use the READMEs, code and recipes; no separate fine-tuning skill installation is required.

Resolve sources

Start with the root README, then the chosen track's README and setup guide. Use managed jobs for supported predefined methods; use interactive sessions for custom training-loop behavior. A custom managed grader alone does not require interactive training.
For managed work, read managed_fine_tuning/README.md, catalog.yml, the selected demo's demo.yaml, README, .env.template, and canonical notebooks/demo.ipynb. Use the learning path/task index to select the demo.
For interactive work, read interactive_training/README.md, docs/auth.md, and the selected recipe's README and configuration.
Interactive model/region eligibility comes from interactive_training/docs/supported_models.md. Managed eligibility comes from Foundry's model-specific documentation and the catalog prerequisites; notebook model roles are not a universal availability guarantee.
Track-specific setup guides define repository paths. Subtree metadata/templates define data and configuration; installed SDK signatures and implementation define operations. Historical changelog entries, benchmarks, and past runtime artifacts do not override these sources. If they conflict, report the conflict and stop before remote operations rather than guessing.

Before updating any README, check whether a more focused guide, recipe, or reference document is a better home for the change. Keep READMEs useful as human-facing entry points: link to detailed material rather than duplicating it or adding unrelated implementation details.

No-spend preflight

Read files and run offline tests, syntax/schema checks, dataset validation, and split/hash checks first. Do not run a whole notebook as a smoke test: later cells may upload data, allocate training, invoke tools, or submit/resume a paid job without another switch.

Read the full file on GitHub · 90 lines

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.

25d ago Changed · +6 lines · -16 tokens per session c9b0cec16f74
28d ago Changed · +10 lines · +133 tokens per session 8ef14274c0af
29d ago First seen · 74 lines · 1,139 tokens per session scan A bbdae5d0ff5f

Subscribe to this mod's changes

fine-tuning AGENTS.md is an instructions file published in the GitHub repository microsoft-foundry/fine-tuning (68 stars, last pushed 2d ago), licensed MIT. It adds 1,256 tokens to every session, about $0.0050 per session on Opus 5.5. 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 2026-10-09.

Data and AI

fine-tuning AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
fine-tuning AGENTS.mdmicrosoft-foundry/fine-tuning	Instructions file	68	2d ago	MIT	A	1,256	Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

fine-tuning: Instructions file for Codex