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/cpuguy96/stepcovnet/agents-mdgit clone --depth 1 https://github.com/cpuguy96/StepCOVNetWrote 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/cpuguy96/stepcovnet/agents-md)<a href="https://agentmods.dev/instructions/cpuguy96/stepcovnet/agents-md"><img src="https://agentmods.dev/badge/instructions/cpuguy96/stepcovnet/agents-md.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 | $0.01611 | $0.01611 |
| Opus 5 | $0.00805 | $0.00805 |
| Sonnet 5 | $0.00322 | $0.00322 |
| Haiku 4.5 | $0.00161 | $0.00161 |
Grade A, and why
StepCOVNet 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 5d 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 — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
StepCOVNet — agent entry
Open one index below. Do not load multiple. Specialized rows (e.g. dataset prep, what's next) are alternatives to the general research/skills index — not extra docs to load together.
| If the task is… | Index |
|---|---|
| A procedure (train, overfit, WSL, log EXP/NOTE, debug playbooks) | .cursor/skills/README.md |
| Research (experiments, design, pipeline; log runs in EXPERIMENT_LOG) | docs/research/README.md |
| Paper draft / citations (related work, BibTeX keys, claim ledger) | docs/research/PAPER_LEDGER.md · paper.bib · PAPER_OUTLINE.md |
DDC placement recreation (Dataset A, M-ddc-20ms) |
configs/ddc/placement_fraxtil.json · scripts/train_ddc_placement.py |
| DDCL / ITGPT recreation (then incremental claims) | PAPER_LEDGER.md · configs/ddc/ddcl_placement_fraxtil.json · configs/ddc/itgpt_placement_fraxtil_exp.json · configs/ddc/itgpt_placement_fraxtil_exp_128ep.json · configs/ddc/itgpt_placement_fraxtil_orig_128ep.json · scripts/train_ddcl_placement.py · scripts/train_itgpt_placement.py |
Onset metrics (timing_match, F1, gates) |
docs/research/ONSET_METRICS.md |
| AR onset design (locked stack, gates, debug notes) | docs/research/AR_ONSET_DESIGN.md |
| Scaling AR training (how to add songs and keep numbers comparable) | docs/research/AR_SCALING_LADDER.md |
| What's next / what to do now / orientation | .cursor/skills/whats-next/SKILL.md → EXPERIMENT_LOG.md § Current phase |
| Research autoresearch (one-prompt goal + budget loop) | .cursor/skills/autoresearch/SKILL.md |
| AR tide overfit (gate PASS — champion v8) | EXPERIMENT_LOG.md § Current phase · champion: configs/ar/tide_overfit.json |
| GPU training, WSL, overfit smoke tests | .cursor/skills/wsl-gpu-stepcovnet/SKILL.md |
Repo layout (src/, scripts/, configs/, data) |
docs/agents/project-layout.md |
Dataset prep / final_data pipeline |
docs/research/DATASET_PREP_PIPELINE.md |
| Recreate training data (SMO downloads → prep) | docs/research/TRAINING_DATA_SETUP.md |
| Pre-push validation (same as CI) | pre_submit.py · .github/workflows/pre-submit.yml |
| Steering correction — how the agent decides / promotes lessons | .cursor/skills/steering-correction-promotion/SKILL.md |
| Refresh agent brain / refresh all rules — audit rules, skills, catalogs | .cursor/skills/agent-brain-refresh/SKILL.md |
North star: PIPELINE_ARCHITECTURE.md.
Rules catalog (always vs scoped — Cursor loads alwaysApply automatically): docs/agents/agent-brain.md.
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.
- 5d ago First seen · 63 lines · 1,611 tokens per session scan A 66ead12b4919
StepCOVNet AGENTS.md is an instructions file published in the GitHub repository cpuguy96/StepCOVNet (22 stars, last pushed 10d ago), licensed Apache-2.0. It adds 1,611 tokens to every session, about $0.0081 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
DeepSpeed AGENTS.md
AGENTS.md instructions for deepspeedai/DeepSpeed, covering deepspeed project rules, commit & ci requirements, code change discipline, test discipline and tool caveats.
streamlit copilot-instructions.md
Copilot instructions for streamlit/streamlit, covering streamlit repo overview, tech stack, folder structure, shell & build policy and make commands.
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.
huggingface_hub AGENTS.md
AGENTS.md instructions for huggingface/huggingface_hub, covering agent guide for huggingfacehub, project overview, setup, key commands and code structure.
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.