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/t-lab/openrfm/agents-mdgit clone --depth 1 https://github.com/T-Lab/OpenRFMWhat 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 | $0.00733 | $0.00733 |
| Opus 5 | $0.00367 | $0.00367 |
| Sonnet 5 | $0.00147 | $0.00147 |
| Haiku 4.5 | $0.00073 | $0.00073 |
Grade A, and why
OpenRFM 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 2d 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 — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository Guidelines
Project Structure & Module Organization
This repository is a KumoRFM-2 reproduction scaffold. The main Python package lives in implementation/kumorfm_repro/, with model code in model.py, synthetic data in data.py, training in train.py, RelBench/file-backed adapters in benchmark_adapter.py and relational_io.py, suite runners in relbench_suite.py and synthetic_suite.py, and target comparison in targets.py. Documentation and status notes live in implementation/README.md and progress_plan/reproduction_status.md. The source paper is in technical_report/. Generated datasets, manifests, checkpoints, and comparisons are under implementation/data/, implementation/runs/, and implementation/targets/; treat these as artifacts, not source modules.
Build, Test, and Development Commands
Work from the implementation directory:
cd /home/user/work/KumoRFM/implementation
python -m compileall kumorfm_repro
Use compileall as the fast syntax/import smoke check. Run a small synthetic training smoke test with:
python -m kumorfm_repro.train --steps 3 --batch-size 2 --context-size 8 --rows-per-child 4 --d-model 32 --layers 1 --output-dir runs/single_smoke
For two-GPU DDP checks:
torchrun --standalone --nproc_per_node=2 -m kumorfm_repro.train --steps 4 --batch-size 4 --context-size 8 --rows-per-child 4 --d-model 32 --layers 1 --amp --output-dir runs/ddp_smoke
For RelBench presets, inspect available suites with python -m kumorfm_repro.relbench_suite --list-presets.
Use benchmark_adapter train-model and relbench_suite as fine-tuning smoke paths only. Paper evidence should come from synthetic pre-training plus icl_eval/icl_suite.
Coding Style & Naming Conventions
Use Python 3 with 4-space indentation, type hints where they clarify interfaces, and descriptive snake_case names for functions, variables, and CLI flags. Keep modules CLI-friendly via python -m kumorfm_repro.<module>. Prefer structured manifests (.json) and parquet/csv table IO over ad hoc text formats. Keep comments short and reserved for non-obvious model, data, or distributed-training behavior.
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.
- 2d ago First seen · 46 lines · 733 tokens per session scan A 4e0d67cdb550
OpenRFM AGENTS.md is an instructions file published in the GitHub repository T-Lab/OpenRFM (22 stars, last pushed 3mo ago), licensed MIT. It adds 733 tokens to every session, about $0.0037 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
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.