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 skills/diyiwuyan/onebench/onebench-deploynpx skills add diyiwuyan/onebench --skill onebench-deploygit 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/skills/diyiwuyan/onebench/onebench-deploy)<a href="https://agentmods.dev/skills/diyiwuyan/onebench/onebench-deploy"><img src="https://agentmods.dev/badge/skills/diyiwuyan/onebench/onebench-deploy.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 5.1 | $0.00087 | $0.02444 |
| Opus 5 | $0.00044 | $0.01222 |
| Sonnet 5 | $0.00017 | $0.00489 |
| Haiku 4.5 | $0.00009 | $0.00244 |
Grade A, and why
onebench-deploy 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
OneBench Deploy
Turn one sentence into the actual OneBench product, not a generic dashboard. Deliver a self-contained HTML file and desktop shortcut first. When phone or multi-device use is requested, keep that local copy and add a user-owned online PWA. Demo, local HTML, PWA, and browser new-tab extension must come from the same OneBench runtime. Preserve the user's name, avatar choice, role theme, module choices, widget order, widget sizes, and data boundary across every output.
Beginner mode
Treat every user as non-technical unless they explicitly request advanced control. Infer the closest pack from their words; do not ask them to learn the catalog. Ask “你希望先管好哪几件事？” only when the request contains no usable role or goal. Do not mention module IDs, GitHub, deployment, tokens, or configuration files before a working local version exists.
Read references/beginner-mode.md before handoff. Give only three plain-language actions: computer opening, phone home-screen installation when requested, and how to say “帮我改成……” next time.
Filling the two blanks
The plain-language values in the starter prompt are valid inputs; do not make beginners learn the catalog first. Normalize them as follows:
- “学生” defaults to the
universitypack (OneBench's student pack means university student); “学习” keeps that pack's course, assignment and certification defaults and includes the learning module. - “K12 教师／老师”、“考研”、“考公”、“内容创作者”、“产品／运营”、“自由职业者” and “团队负责人” map to their identically named first-party packs.
- Use the 1–3 things after “最想管理” to prioritize the default modules and title. A broad word such as “学习” is sufficient for a working first version; a more concrete list improves the result but is never required.
- Use the role pack's theme by default. If the user gives a preferred color or style, keep the role modules and change only the theme.
- If the user provides a name or preferred form of address, write it to
workspace.profile.displayName; otherwise use a warm generic address and let them change it in “定制”. - Map “国考／省考／事业编／遴选／行测／申论／考公冲刺台” to the
examprofessional edition andcivil-service-exampack. Map “班主任／小学老师／初高中老师／任课教师／教务／带班” to theteacherprofessional edition andteacherpack. Map “胡楚靓同款” and “创作者工作台／创作者专业版” tohuandcreator. Do not imitate any of these with a recolored basic pack. - Treat a professional edition as a portable product state: carry
editionandprofessionalDatain the downloaded standalone HTML; usepublic/onebench-seed.jsonfor the first open of a user-owned online repository. Never deliver a professional-looking page that reopens as the basic version.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 91 lines · 87 tokens per session scan A a0464b37d58c
onebench-deploy is a skill published in the GitHub repository diyiwuyan/onebench (18 stars, last pushed today), licensed MIT. It adds 87 tokens to every session and 2,444 once invoked, about $0.0004 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 skills, from other repositories
copilotkit-channels
Use for the CODE half of a managed Intelligence Channel with Slack or Microsoft Teams: customising the Channel a CLI-scaffolded project already ships, or — for a project the CLI did not generate — writing the Channel declaration, the long-running host, and the awaited activation call. Teams provider setup is in scope…
setup-slack-channel
Use for the PROVIDER half of getting a locally running CopilotKit Channels agent to answer in Slack, when no Slack app exists yet — setting up a Channels bot in Slack for the first time, creating the Slack app and its tokens, attaching it to a managed Intelligence Channel, or when a Channel reports setuprequired, sits…
channels-setup
Use when a developer wants to build their first CopilotKit Channels agent and get it answering in Slack or Microsoft Teams — "set up a channel", "connect my agent to Slack", "get my agent into Teams", or starting from nothing and wanting a working channel end to end. Covers the whole path: inspecting or scaffolding…
product-update-newsletter
Draft, update, or audit a crisp, changelog-grounded Anarlog product-update newsletter in Loops (app.loops.so) for a desktop release. Use after the changelog is merged, when asked to draft, revise, or pre-send check the release announcement email.
anarlog
Query Anarlog meetings, notes, summaries, transcripts, participants, action items, and recurring history. Use when a user asks about their Anarlog meeting data or needs meeting context for another task.
gmail
Manage Gmail email — drafting, sending, organizing, filters, vacation replies, and inbox analysis.