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/deepspeedai/deepspeed/agents-mdgit clone --depth 1 https://github.com/deepspeedai/DeepSpeedWhat 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 | $0.00455 | $0.00455 |
| Opus 5 | $0.00228 | $0.00228 |
| Sonnet 5 | $0.00091 | $0.00091 |
| Haiku 4.5 | $0.00046 | $0.00046 |
Grade A, and why
DeepSpeed 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 yesterday.
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.
Copies of this mod
1 near-identical copy found in the catalogue:
- DeepSpeed CLAUDE.md — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 33 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — Workspace-level instructions for AI coding agents
DeepSpeed Project Rules
Commit & CI requirements
- All non-merge commits MUST have a
Signed-off-byline (use--signoff). Get the name and email fromgit config user.name/git config user.email. - Formatting: yapf (column_limit=119,
.style.yapf) + flake8 (.flake8). - Always verify changed files pass pre-commit checks before committing:
pre-commit run --files <changed_files>. Only check modified files, not the entire codebase. Config:.pre-commit-config.yaml. check-torchdisthook: NEVER directly import torch's distributed module. Useimport deepspeed.comm as distinstead.- New files require license header:
# SPDX-License-Identifier: Apache-2.0 # DeepSpeed Team
Code change discipline
- NEVER make cosmetic/formatting-only changes to existing code. Only add/modify lines that are functionally necessary. Minimizing diff noise is critical for code review.
- Delete dead code decisively — if code is unused at runtime (only referenced in tests), remove it along with its tests.
- Prefer consolidating tests over proliferating test files.
- Blend in: when modifying code, read the surrounding context and match the style of neighboring code (naming, spacing, patterns, idioms).
- Write beginner-friendly code: avoid deeply nested expressions or chained logic. Break complex expressions into clear, named intermediate steps.
- Comments should explain why, not what. Describe the purpose and reasoning, not the mechanics that the code already shows.
- New features must include corresponding tests and documentation updates.
Tool Caveats
Edit tool auto-formatter
The Edit tool has a hidden auto-formatter that silently changes quotes, whitespace, blank lines, and line wrapping. For format-sensitive modifications (e.g., when exact formatting matters for pre-commit), use bash with sed, python, or cat instead.
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.
- yesterday First seen · 33 lines · 455 tokens per session scan A 78604afb71eb
DeepSpeed AGENTS.md is an instructions file published in the GitHub repository deepspeedai/DeepSpeed (43,041 stars, last pushed yesterday), licensed Apache-2.0. It adds 455 tokens to every session, about $0.0023 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 2026-09-01.
Other instructions, from other repositories
skypilot AGENTS.md
AGENTS.md instructions for skypilot-org/skypilot, covering claude.md - skypilot development guide, project overview, repository structure, development setup and environment setup.
pytorch CLAUDE.md
Claude Code instructions for pytorch/pytorch, covering ai policy — mandatory, scratch space, pr review, environment and ci docker images.
pytorch copilot-instructions.md
Copilot instructions for pytorch/pytorch, covering pytorch copilot instructions, architecture overview, core components, the code generation workflow and development workflows.
nvcf AGENTS.md
Instructions for NVIDIA/nvcf, covering agents.md - guide for ai coding agents, repo layout, cross-repo and stack routing, oss snapshot hygiene and local qa and testing environment safety.
pytorch AGENTS.md
AGENTS.md instructions for pytorch/pytorch, a project described as: Tensors and Dynamic neural networks in Python with strong GPU acceleration.
auto-deep-researcher-24x7 AGENTS.md
Instructions for Xiangyue-Zhang/auto-deep-researcher-24x7, covering agents.md — project-level rules for ai assistants, 🚨 contributor policy — non-negotiable, hard rules, required pre-push verification and how to commit properly.