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/onescience-ai/oneskills/onescience-trainernpx skills add onescience-ai/OneSkills --skill onescience-trainergit clone --depth 1 https://github.com/onescience-ai/OneSkillsWrote 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/onescience-ai/oneskills/onescience-trainer)<a href="https://agentmods.dev/skills/onescience-ai/oneskills/onescience-trainer"><img src="https://agentmods.dev/badge/skills/onescience-ai/oneskills/onescience-trainer.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 | $0.00147 | $0.02941 |
| Opus 5 | $0.00073 | $0.01470 |
| Sonnet 5 | $0.00029 | $0.00588 |
| Haiku 4.5 | $0.00015 | $0.00294 |
Grade A, and why
onescience-trainer 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 — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
输入获取方式
本技能支持两种输入方式:
- 上下文 handoff(默认):从调用方传入的
step_handoff获取任务信息。 - 文件 handoff(autonomous_mode):从
.onescience/handoff/step_{step_id}.yaml读取任务信息。执行后,将结果写入.onescience/handoff/step_{step_id}_result.yaml。
启动时优先检查 .onescience/handoff/ 目录是否存在对应的交接文件;若存在则使用文件模式,否则使用上下文模式。
文件交接格式参见 skills/onescience-orchestrator/references/file_handoff_contract.md。
OneScience Trainer
职责
将本技能作为科学模型训练任务的执行工作流。把 orchestrator 交给本技能的当前步骤,例如“根据网页说明生成训练脚本”“基于已有 checkpoint 继续训练”“按文本给定的数据与优化器约束生成微调入口”“整理训练计划并提交训练”,转化为从训练信息获取到结果验证的可追踪闭环。
本技能只消费当前执行步骤,不重新规划完整用户目标。本技能负责训练工作流契约和阶段顺序,负责明确训练策略、数据配置、split 假设、loss、optimizer、scheduler、评测频率、checkpoint 策略、日志与验证要求等完整训练脚本内容;这些内容由 trainer 直接定义并保留在 trainer_workdir 中。训练执行阶段的外层控制、入口确定、预检要求和结果归因也由 trainer 负责。只有当这些已定义内容需要写入仓库、修改现有项目文件或对接项目原生目录结构时,才把当前步骤中的文件落盘子动作交接给 onescience-coder。依赖包、硬件和运行环境需求不单独成阶段,而是在训练执行阶段通过 onescience-runtime 统一处理。
工作流
创建或更新工作目录;当 step_handoff.inputs.runtime.code_save_dir、上游交接物或用户明确指定了代码保存目录时,trainer_workdir 必须使用 <code_save_dir>/.trainer_work/<run_id>/。其中 code_save_dir 用于保存最终训练脚本、配置、日志引用和用户要求的结果输出,trainer_workdir 用于保存训练知识、计划、manifest、runtime 请求/结果、验证报告等 trainer 中间知识产物。若上游已显式提供 step_handoff.inputs.runtime.trainer_workdir,则应直接使用该目录,并要求其与 code_save_dir 语义保持一致;若未提供代码保存目录,但 step_handoff.inputs.runtime.workdir、task_context.relevant_artifacts 或用户明确指定了工作目录,则沿用该目录;否则兼容性回退到 .onescience/trainer/<run_id>/。进入具体代码生成或执行阶段时,必须从 trainer_workdir 中已保存的知识文件读取并交接,不依赖未落盘的会话上下文。
本技能不依赖 scripts/ 目录下的预置脚本;训练入口、配置、数据适配器、损失函数、评估逻辑和测试要求等完整训练脚本内容都由 trainer 先行定义,不再把这些内容重新交给 onescience-coder 决定。若需要将这些已定义内容写入仓库、修改现有项目文件或对接项目原生目录结构,再交接给 onescience-coder 落盘为明确文件。
在消费上游交接或生成产物前,先读取 references/workflow_contract.md。随后必须先根据 step_handoff.step_goal、task_context.user_goal、inputs、resource_bindings 和用户直接请求,从 type=resource 技能中召回与训练目标相关的资源,获取训练规格知识、模型/数据使用知识和训练规划决策知识;无论上游是否提供了 resource_bindings,都不能跳过初始资源召回。资源 path 仅用于标识和追踪,允许作为训练依据的只有用户明确提供内容、上游已展开的资源内容和资源技能返回的 matched_resources[*].content。
What ships with it
8 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.
- references/phase_0_input_acquisition.md 3.7 KB
- references/phase_1_training_knowledge.md 3.6 KB
- references/phase_2_data_and_split.md 4.4 KB
- references/phase_3_training_strategy.md 6.5 KB
- references/phase_4_codegen.md 5.7 KB
- references/phase_5_execution.md 3.9 KB
- references/phase_6_validation.md 2.5 KB
- references/workflow_contract.md 12 KB
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 · 139 lines · 147 tokens per session scan A 9b179b2bb634
onescience-trainer is a skill published in the GitHub repository onescience-ai/OneSkills (20 stars, last pushed 21d ago), licensed MIT. It adds 147 tokens to every session and 2,941 once invoked, about $0.0007 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
nsfc-budget
当用户明确要求“写/生成 NSFC 预算说明书”“写预算说明”“生成 budget.tex / budget.pdf”“写国自然预算 justification”时使用。基于用户标书正文或补充材料,输出一份可提交的预算说明书 LaTeX 项目并渲染 budget.pdf。若用户未指定工作目录,必须暂停并先要求其指定。⚠️ 不适用:用户只是想了解预算原则;用户仅要预算表数字而不写说明书;或用户是 2026 青年 A/B/C 默认包干制且无需预算说明书的场景。.
paper-select-journal
当用户明确要求“推荐投稿期刊”“帮我的论文选 SCI 杂志”“这篇 manuscript 适合投哪些 journal”“期刊匹配/选刊/投稿建议”时必须使用。适用于用户提供全文、摘要、Markdown、LaTeX、PDF、Word 或混合材料的场景;本 skill 会基于 manuscript 与用户偏好,先用内置 2023IF.xlsx 做最小硬过滤生成候选池,再由宿主模型自主规划 Set1/Set2/Set3,并联网核验 scope / 质量 / 近 3 个月 PubMed 论文,最后输出 1 份按推荐度排序的 Markdown 选刊报告。⚠️…
paper-search
Search, download, and read academic papers from 20+ sources (arXiv, PubMed, Semantic Scholar, CrossRef, etc). Use when the user asks to find papers, search for research, look up academic literature, download a paper PDF, or extract text from a paper.
paper-planning
Guides pre-writing planning for academic papers with 4 structured steps: story design (task-challenge-insight-contribution-advantage), experiment planning (comparisons + ablations), figure design (pipeline + teaser), and 4-week timeline management. Includes counterintuitive planning tactics (write a mock rejection…
evo-memory
Manages persistent research memory across ideation and experimentation cycles. Maintains two stores: Ideation Memory MI (feasible/unsuccessful directions) and Experimentation Memory ME (reusable strategies for data processing, model training, architecture, debugging). Three evolution mechanisms: IDE (after…
paper-review
Guides self-review of YOUR OWN academic paper before submission with adversarial stress-testing. Core method: 5-aspect checklist (contribution sufficiency, writing clarity, results quality, testing completeness, method design), counterintuitive protocol (reject-first simulation, delete unsupported claims, score trust…