Training attestation — reproducible, attributable, auditable off-site standardized training for asset pricing. Hard-gates four failure modes (half-refresh, hash mismatch, unknown-read-as-pass, master pollution). Ships one skill.
Install
# from GitHub (first run asks for allowBuilds approval — follow the hint, retry)
dsh plugin --profile web add github:Z-Asset/attest
Any plugin you install runs third-party code with your own permissions — it can read your files, use your credentials, and reach the network, and tool approvals don’t sandbox it. GitHub-sourced plugins also run build scripts at install time — pnpm blocks those until you allow them, so an install can stop with ERR_PNPM_GIT_DEP_PREPARE_NOT_ALLOWED or ERR_PNPM_IGNORED_BUILDS; dsh prints the exact key to add under allowBuilds in your profile’s pnpm-workspace.yaml, and the install works on the next run. Allowing a build is a trust decision: only install sources you trust, and pin a commit (github:owner/repo#sha).
README
This plugin publishes its README in Chinese only.
把可移动三件套(X/ + Y/ + train_from_xy.py)的异地标准化训练,钉成一条可复算、可归因、可审计的裁决链。
它是什么
attest 不训练模型、不改一行策略代码。它做的事,是让一次训练可信到可以上线:
- 把 Z 母本包封成
zip + DISPATCH.json(sha1 实算,不抄声明)发到异地; - 异地(只有 Jupyter、操作者不懂内部细节)跑
doctor → train → eval → collect; - 结果封回本机,由人独断 promote。
四道闸硬拦四类事故:
| 闸 | 拦的事故 |
|---|---|
| 锚点一致性 | 半拉子刷新(MANIFEST 与 Y_meta 不是同一天) |
| sha1 账物对账 | 账物不符(声明哈希与实算不符) |
| 三态口径 | 「判不了」被读成「通过」 |
| 母本守卫 | 手滑污染 Z/D 线上目录 |
铁律:「判不了」绝不等于「通过」。这是全套防呆里最容易被人绕过去的一条。
结构(双轨单源)
本包同时承载两套结构,SKILL.md 只维护一份(plugins/attest/skills/):
- Claude Code 插件 —
.claude-plugin/marketplace.json+plugins/ - DSH/npm 插件 —
package.json+cordis.patch.yml+src/index.ts
安装
dsh plugin --profile <name> add @z-asset/attest
用法
异地侧五个动词(详见 plugins/attest/skills/attest/README_FIRST.md):
python pt.py doctor --bundle <包目录> # 体检,不通过就停
python pt.py train --bundle <包目录> --mode smoke # CPU 冒烟
python pt.py train --bundle <包目录> --mode week --commit # 真训练(要 GPU)
python pt.py eval --bundle <包目录> --workdir <空目录> # 评估与裁决
python pt.py collect --workdir <空目录> --out . # 打包回传
本机侧打包发出:
python scripts/pt_pack.py pack --bundle <Z包目录> --out <本地目录>
边界
不写 Z 母本、不改任何策略包、不改 _sync 生成器、不改 D:\ZML\trading\*。promote 只在本机、由人独断。
Links
More in this category
loopx-project/loopx#dsh-loopx-plugin★ 6398
LoopX, a provider-neutral, local-first state kernel and control plane for long-horizon agents: keeps Goal, Todo, gate, evidence, quota, recovery, and handoff state above DeepSeek Harness, while the plugin bootstraps the CLI and skills, admits bounded same-session continuation, and adds a loopback GoalBar for the exact bound loop.
Q00/ouroboros#integrations/dsh-plugin★ 6260
Config-only bundle that mounts Ouroboros through the DSH MCP client, exposing 36 interview, Seed, execution, evaluation, and evolution workflow tools in DSH.
chuspeeism/dashi-taskboard#deepseek-harness★ 3388
Embeds the active installed Codex Taskboard runtime in the DeepSeek Harness sidebar, using its launcher runtime descriptor instead of a fixed port.
NanmiCoder/dsh-agent-teams★ 2368
AgentTeams multi-agent teams.
EthanYoQ/AI-Novel-Writer#dsh-ai-novel-writer★ 1753
Installs a dedicated AI novel-writing preset and workbench: revisioned local project assets, a compact side drawer, and native approval-gated single-file changes.
Miaotofu01/Study-Mate★ 1144
Your AI learning companion: helping you plan your studies, understand concepts, and build projects to master a subject.
Community comments
Comments are public GitHub Discussions. Loading them connects to GitHub and Giscus; a GitHub account is required to post.