MIT-licensed, dsh-native Chinese resume clinic — a `resume_matcher` archive tool plus the hr-critic five-step audit skill, wired to the bundled a4cv editor for layout tweaks and PDF export, with no API key and no Python dependency.
DSH integration
Native runtime
Author-claimed
Safety audit
Unaudited
Last verified
2026-11-15
License
MIT
01What can it help you accomplish?
Turn a resume file plus a target job description into an HR-style audit and a rewritten resume
A five-step audit report (first impression, ground-level audit, revision blueprint, rewritten resume, final verdict) with the report and the optimized resume archived as JSON
Chinese-speaking job seekers and career switchers tailoring a resume to one specific role
Fine-tune the rewritten resume's layout in the bundled a4cv editor and export a PDF without leaving dsh
An a4cv editing session with 34 style presets, layout / color / font-size tweaks and PDF export; the optimized resume can also be copied or downloaded
Candidates who need a print-ready PDF, plus reviewers browsing the stored resumes, JDs and reports in the sidebar panel
02How to install into DeepSeek Harness
Prerequisites
- dsh 0.2.x — the README's install block is labelled for dsh 0.2.x and says to clone the repo locally and install it as a link
- A text-based PDF or DOCX resume — `.doc` / `.pages` and scanned-image PDFs are not supported
Installation steps
- 01
Run the README's install command: `dsh plugin add https://github.com/liangdabiao/dsh-resume-matcher`
$ dsh plugin add https://github.com/liangdabiao/dsh-resume-matcher
- 02
On dsh 0.2.x, clone the repository to your machine first and install it as a link, as the install comment in the README notes
- 03
Let the dependencies install automatically — `@deepseek-ai/schemastery` and `unpdf` are pulled in by the install
Verify the integration
- The sidebar shows the plugin's 「简历诊所」panel with 概览 / 简历库 / 岗位库 / 分析报告 / 发起分析 views — that panel is the UI the README lists for this plugin
- Plugin data is written under `~/.dsh/resume-matcher/`, the data directory the README documents, which `DSH_HOME` relocates
03DSH integration and capability boundaries
Native dsh plugin: registers the `resume_matcher` tool, provides the hr-critic skill, injects system-prompt rules and mounts a 「简历诊所」sidebar panel
`resume_matcher` archive tool
A PDF / DOCX resume file path in the conversation, or pasted resume text→A structured JSON resume archive; double-column DOCX layouts are parsed with their order preserved
writes JSON files under `~/.dsh/resume-matcher/` (resumes / jobs / reports)hr-critic five-step audit skill
An archived resume plus a target job description→A five-step report: first impression, ground-level audit (critique / analysis / suggestion), revision blueprint, rewritten resume, final verdict
a4cv visual editor hand-off
The Step 4 optimized resume markdown extracted from a stored report→An a4cv session with 34 style presets, layout / color / font-size tweaks and PDF export
hands the resume over through same-origin sessionStorage (`?pickup=session`), with an extra new-window `?md=` channelthe bundled a4cv static assets are integrated unchanged「简历诊所」sidebar panel
Stored resumes, job descriptions and analysis reports→Overview / resume library / job library / analysis reports / start-analysis views, where start-analysis opens a new dsh conversation pre-filled with a draft
the panel follows the dsh light / dark theme
04Who is it for? When not to use it?
Good for
- Chinese-speaking job seekers and career switchers tailoring a resume to one specific role
- Candidates who need a print-ready PDF, plus reviewers browsing the stored resumes, JDs and reports in the sidebar panel
Not for
- Only text-based PDF and DOCX resumes are parsed. `.doc`, `.pages` and scanned-image PDFs are unsupported and must be converted to a text-based PDF or pasted as text.
- Audit conclusions are generated by the model and are professional opinion rather than a factual guarantee; the README explicitly asks users not to fabricate experience on their resume.
05Compatibility, maintenance and safety notes
- Only text-based PDF and DOCX resumes are parsed. `.doc`, `.pages` and scanned-image PDFs are unsupported and must be converted to a text-based PDF or pasted as text.
- Audit conclusions are generated by the model and are professional opinion rather than a factual guarantee; the README explicitly asks users not to fabricate experience on their resume.
- a4cv's built-in 「AI 优化」 needs a model API configured inside the editor; that is independent of the plugin itself, which runs with zero API key.
- v2 already ships the a4cv integration; a built-in OpenAI-compatible client (re-run from the panel) is only listed as optional future work.
MIT · no published release · last push 2026-11-13
06Frequently asked questions
How do I install dsh-resume-matcher in DeepSeek Harness?
On dsh 0.2.x run `dsh plugin add https://github.com/liangdabiao/dsh-resume-matcher`; the README's install comment also says to clone the repo locally and install it as a link. Dependencies (`@deepseek-ai/schemastery`, `unpdf`) are installed automatically, and data is stored in `~/.dsh/resume-matcher/`, relocatable with `DSH_HOME`.
Do I need my own model API key for the audit?
No. The README frames the plugin as skill-driven with zero API key and zero Python dependency — the audit is executed by the session model of your dsh conversation through the hr-critic skill. Only a4cv's built-in 「AI 优化」 needs a model API configured inside the editor, and that is unrelated to the plugin itself.
Which resume formats can it read?
Text-based PDF and DOCX, including double-column DOCX layouts, which are parsed with their order preserved. `.doc`, `.pages` and scanned-image PDFs are not supported — convert them to a text-based PDF or paste the text into the conversation.
Where does the PDF export happen?
In the bundled a4cv editor: the report page's 「编辑器微调」 entry opens the Step 4 rewritten resume there, where you choose one of 34 style presets, adjust layout / color / font size and export a PDF. The optimized resume can also be copied or downloaded, and the hand-off runs over same-origin sessionStorage (`?pickup=session`) with an extra new-window `?md=` channel.
What does the sidebar panel add on top of the chat flow?
The 「简历诊所」panel gives overview / resume library / job library / analysis reports / start-analysis views, its theme follows dsh light and dark, and start-analysis opens a new dsh conversation pre-filled with a draft for the chosen resume × job pair.
07Related DSH workflows
dsh-univer-office
by dream-num
Give DeepSeek Harness a real office environment. Univer Office Plugin brings spreadsheets, docs, slides, canvases, relational tables, and more into one runtime — with connected data, validation, versioned changes, and isolated worktrees for multi-agent collaboration.
clearai-dsh
by clearailhc
Brings an evidence-checked epistemic loop to DSH: research sessions settle into a browsable domain ontology with evidence chains and support levels; contradictory readings are surfaced for human adjudication.
dsh-fish-memory
by yuuyuko-uu
Stores time-stamped, decaying per-session memories, recalls relevant items, and optionally distills imported local archives through the configured DSH model.
dsh-trading
by zhu1090093659
Agent-native trading terminal built on DeepSeek Harness. Crypto, US, CN and HK in one three-column GUI, 19+ hot-swappable connectors, dry-run by default with human approval on every live order. BYOK, no data redistribution.
08Data and sources
DeepSeek Harness(dsh)插件:HR 视角的中文简历深度优化工作台。
Host 半边:resume_matcher 工具 + 技能提供方 + 系统提示注入 + 面板路由
本插件为其 dsh 化形态:**纯技能驱动、零 API key、零 Python 依赖**
This page is generated from the project’s public documentation, repository metadata and a structured parse of DSH Plugins; last verified on 2026-11-15. Found an error? Submit a correction.
Best DeepSeek Harness Plugins
Twelve plugins worth installing first — picked from the whole catalog, across every category.
