DeepSeek Harness Plugin Hub

Publish and manage complete Harness Profiles. Discover Plugins for your next setup.

Explore

PluginsPresetsDocsNews

Community

Publish a pluginContactReport an issue

Resources

Plugin Hub on GitHubDeepSeek HarnessSystem statusPrivacy notice
© 2026 DeepSeek Harness Plugin HubPowered byPaxTech

Independent and unofficial. Not affiliated with, authorized by, or endorsed by DeepSeek.

Llm As A Verifier — DSH Plugin for DeepSeek Harness
← Plugins
L

dsh-llm-as-a-verifier

Llm As A Verifier

LLM-as-a-Verifier for DeepSeek Harness: fine-grained reward, Probabilistic Pivot Tournament best-of-N selection, and per-step progress tracking as agent tools.

The plugin will be installed here. Keep web if you are unsure.

npx -y @deepseek-ai/dsh plugin --profile web add github:TaurenMountain/dsh-llm-as-a-verifier#c2727fe375e7900f490730d476b08b887ac8d6f2
READMECompatibilityVersions

Compatibility and provenance

Llm As A Verifier is published as dsh-llm-as-a-verifier and currently resolves to version 0.1.1. The Hub verifies its manifest and preserves the exact installation source for reproducible installs.

DSH compatibility
*
Runtime surfaces
any
Release source
github
Registry updated
8/20/2026

Versions

0.1.1stable
8/20/2026

README

dsh-llm-verifier

English | 中文

npm 包名:dsh-llm-as-a-verifier(裸名 dsh-llm-verifier 在 npm 上已被他人占用,仓库与 npm 包统一为 dsh-llm-as-a-verifier)。

把 LLM-as-a-Verifier(arXiv 上的统一验证框架)魔改进 DeepSeek Harness(dsh):让智能体在干活时能对自己的候选答案做细粒度概率化验证——不再是「好/坏」一锤子判断,而是读取验证模型在 20 级评分字母上的完整 logprob 分布并取期望。

你会得到

三个模型可直接调用的工具(安装后自动注册,无需重启即可用):

工具能力复杂度
verify_compare对两个候选(代码/方案/轨迹)按评价标准打分,返回 [0,1] 的细粒度奖励 (scoreA, scoreB)1 次验证调用 / 标准 / 次
verify_selectN 选一:Probabilistic Pivot Tournament(概率枢纽锦标赛),O(Nk) 次比较替代 O(N²) 全循环线性于 N
verify_track逐步进度追踪:验证器以 A(0%)..T(100%) 为每个 checkpoint 打分,画出进度曲线O(K) 次调用

为什么比「LLM-as-a-Judge」更细? 上游框架的核心思想是:① 用细粒度评分(20 级字母尺度);② 对评分 token 的完整 logprob 分布取期望(而不是只取 argmax);③ 用重复评估 + 标准分解缩放可靠性。本插件原样移植了这套逻辑(打分提取、成对提示词、锦标赛、进度追踪、token 计量),并适配为 DSH 的 Cordis 工具插件。

安装

dsh plugin --profile web add dsh-llm-as-a-verifier

要求:dsh ≥ 0.1.0-rc.6、Node ≥ 18。安装完成后重启 dsh web(或等待 HMR 自动生效)。

配置验证后端

验证模型必须是能返回 token 级 logprobs 的 OpenAI 兼容服务:DeepSeek 官方 API、vLLM/SGLang 本地服务、OpenAI 等均可。

在你的 profile 配置(~/.dsh/profiles/<name>/cordis.patch.yml 或 ~/.dsh/cordis.patch.yml)里写:

- id: llm-verifier
  config:
    baseUrl: https://api.deepseek.com   # 或 vLLM: http://localhost:8000/v1
    apiKey: '${DEEPSEEK_API_KEY}'       # 推荐用环境变量,见下
    model: deepseek-v4-flash            # 不填时:DeepSeek 默认 deepseek-v4-flash,其余自动探测 /models
    maxConcurrency: 8

凭证解析顺序(与上游一致):插件 config → OPENAI_BASE_URL + OPENAI_API_KEY → DEEPSEEK_API_KEY(自动启用 DeepSeek 端点与 thinking 参数)。什么都不配时,工具注册不受影响,调用时才报 MissingAPIKeyError。

export DEEPSEEK_API_KEY=sk-...   # 最省事的配置方式

使用示例

装好后直接在对话里让智能体用(无需额外命令):

我写了三个候选实现,帮我用 verify_select 按「正确性、性能」标准选出最好的,
然后对选中的实现用 verify_track 检查我之前的修复步骤是否有进展。

或者手动指定:

verify_compare: problem="写一个反转字符串的函数", candidateA="def rev(s): return s[::-1]",
candidateB="def rev(s): return s", criteria={"Correctness": "代码是否真的反转了字符串?"}

配置项

配置默认说明
modelDeepSeek: deepseek-v4-flash;其余自动探测验证模型名
baseUrl按凭证推断OpenAI 兼容端点
apiKey按环境推断建议走环境变量
timeoutMs60000单次请求超时(毫秒)
maxConcurrency8最大并发验证调用
deepseek按 baseUrl 推断强制 DeepSeek 调用路径(thinking 开启)
prefilltrue非 DeepSeek 服务器上对评分标签做 prefill(vLLM/SGLang 读取字母分布更稳)
compare / select / tracktrue是否注册对应工具

工具参数(nEvaluations 重复评估次数、pivots 枢纽数、seed 环赛种子、groundTruthNote 基准提示等)与上游 llm_verifier Python 包一一对应,详见 使用手册。

作为库使用

import { Verifier } from 'dsh-llm-as-a-verifier'

const verifier = new Verifier({ baseUrl: 'http://localhost:8000/v1' })
const { scoreA, scoreB } = await verifier.compare(problem, a, b, { Correctness: '...' })
const result = await verifier.select(problem, candidates, { Correctness: '...' }, { pivots: 2 })
const curve = await verifier.track(problem, steps, { checkpoints: [1, 3] })

开发与测试

npm ci
npm run check   # typecheck + vitest(76 个用例,含本地 mock logprobs 服务器端到端测试)
npm run build

TDD 过程与全流程 SOP 见 docs/SOP.md。

许可与致谢

MIT。打分期望、成对提示词、Probabilistic Pivot Tournament、进度追踪与 logprob 提取逻辑移植自 llm-as-a-verifier/llm-as-a-verifier(MIT),完整归属见 LICENSE。

Related plugins

More verified plugins in agents-orchestration.

Headless@deepseek-ai/dsh-headlessThe dsh one-shot bundle: a direct core Agent/Session runner over dsh-base with no Host, HTTP, or browser layerExperimental Agent Team Web Profile@deepseek-ai/dsh-experimental-agent-team-web-profileExperimental Web profile layer for Agent Teams Remote and UI pluginsSubagent Codex@deepseek-ai/dsh-subagent-codexOne-shot Codex subagent provider over the official app-server protocolSubagent Claude Code@deepseek-ai/dsh-subagent-claude-codeOne-shot Claude Code subagent provider over the official Agent SDK
Latest
0.1.1
DSH
*
HMR
Process restart
Tree shaking
Safe tree shaking not declared
Unpacked size
Unavailable
Files
Unavailable
Surface
any
License
MIT
Source
github
GitHub
★ 6
Weekly downloads
0
Last push
8/24/2026
View source ↗
README badge

Click the badge to copy Markdown for your README.

Do you maintain this Plugin?Claim benefit · Priority security scan

Verify the GitHub repository declared in package.json to manage this listing. After you claim it, Hub will prioritize a security scan of the current version and publish the result when it passes.

Claim this Plugin →
Report an issue
DeepSeek Harness Plugin Hub
ProfilesPluginsCategoriesNewsDocsSign inManage Profiles
ProfilesPluginsCategoriesNewsDocsSign in