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/diyiwuyan/onebench/agents-mdgit clone --depth 1 https://github.com/diyiwuyan/onebenchWrote 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/diyiwuyan/onebench/agents-md)<a href="https://agentmods.dev/instructions/diyiwuyan/onebench/agents-md"><img src="https://agentmods.dev/badge/instructions/diyiwuyan/onebench/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 | $0.00723 | $0.00723 |
| Opus 5 | $0.00362 | $0.00362 |
| Sonnet 5 | $0.00145 | $0.00145 |
| Haiku 4.5 | $0.00072 | $0.00072 |
Grade A, and why
onebench 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 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 — 28 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Prototype Instructions
Run the local server yourself and open the preview in the browser available to this environment. Do not give the user server-start instructions when you can run it.
Before making substantial visual changes, use the Product Design plugin's get-context skill when the visual source is unclear or no longer matches the current goal. When the user gives durable prototype-specific design feedback, preferences, or decisions, record them in AGENTS.md.
When implementing from a selected generated mock, treat that image as the source of truth for layout, component anatomy, density, spacing, color, typography, visible content, and hierarchy.
Build app UI in src/. Keep .openai/hosting.json, worker/index.js, scripts/prepare-sites-build.mjs, and tests/sites-worker.test.mjs intact so the same local prototype can be handed to Sites. Before a Sites handoff, run npm run build and npm run test:sites; the build must leave dist/client/index.html, dist/server/index.js, and dist/.openai/hosting.json.
Durable OneBench product direction
-
The lifestyle flagship is held to a six-screen reference-matrix standard. Do not call it complete because it has similarly named modules; each captured reference screen must have a matching OneBench route, hierarchy, primary controls, persisted state, and browser QA evidence before release.
-
The actual daily workbench is the product. Studio, onboarding, module management, sync, and help belong in secondary panels instead of dominating the home screen.
-
Default visual direction: warm paper texture, calm lifestyle feeling, strong role identity, mobile-first hierarchy, and a denser responsive desktop layout. Avoid generic dark admin dashboards.
-
Every visible module must be usable. Never represent a promised feature only as an English module ID or an “installed” badge.
-
The Demo, downloaded offline HTML, PWA, and browser new-tab extension must run the same app code and show the same workbench. Do not maintain a reduced export-only dashboard.
-
First use should already contain realistic role-specific sample content so a beginner understands what to do without configuring anything.
-
Keep technical concepts such as GitHub tokens, repository paths, registries, and update sources out of the default beginner flow.
-
Keep the homepage intentionally selective: use the left rail for app/module entrances and the home area for a user-curated widget canvas.
-
Homepage widgets must support persistent reordering and size changes. Role-pack sample content is only a first-use seed; every visible item must later be editable or removable by the user.
-
Treat calendar and weather as first-class shared widgets. Weather may refresh from the network but must keep a useful local cache and never block the offline workbench.
-
Community contributions are not one generic “template”: distinguish career packs, workspace layouts, theme packs, module bundles, and individual modules.
-
Professional editions must not expose a version switcher in the main top bar. Put switching, returning to the basic edition, profile/display controls, and data backup in the bottom-left Settings entry.
-
A professional edition is a distinct workflow and data model, not a recolored shared page. Exam, teacher, lifestyle/creator-reference, and content-creator editions need role-specific create/update/delete flows and independently persisted data.
-
Every edition is a first-class deliverable: its Settings must expose a beginner-safe local download, the exported HTML must carry the chosen edition and its professional data, and a user-owned online repository must first-open the selected edition through
public/onebench-seed.json.
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 · 28 lines · 723 tokens per session scan A c3e19e24d3a1
onebench AGENTS.md is an instructions file published in the GitHub repository diyiwuyan/onebench (18 stars, last pushed 4d ago), licensed MIT. It adds 723 tokens to every session, about $0.0036 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
tutor-management-app CLAUDE.md
Instructions for t1112000/tutor-management-app, covering claude.md, commands, architecture, auth and page pattern.
tutor-management-app AGENTS.md
Instructions for t1112000/tutor-management-app, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
openscreen AGENTS.md
AGENTS.md instructions for getopenscreen/openscreen, covering agents.md, setup commands, development principles, project layout and code style.
Claw3D AGENTS.md
Instructions for iamlukethedev/Claw3D, covering agent instructions, cursor cloud specific instructions, service overview, running the app and lint, typecheck, and tests.
commonly CLAUDE.md
Claude Code instructions for Team-Commonly/commonly, covering claude.md / agents.md, 🧠 product vision & architecture philosophy, what commonly is, the architecture model and key concepts.
claude-ads CLAUDE.md
Instructions for AgriciDaniel/claude-ads, covering claude ads repository instructions, architecture, development and verification.