MIT-licensed skill bundle for DeepSeek Harness: academic paper revision, rebuttal preparation and presentation storyboarding with packaged evidence, diff-level edits and validation reports.
DSH integration
Native runtime
Author-claimed
Safety audit
Unaudited
Last verified
2026-11-11
License
MIT
01What can it help you accomplish?
Revise an academic manuscript against supplied experiment records and references
Traceable suggestions, the exact edits as a diff, a revised manuscript, and a checks report — written to a run directory
Researchers and students who want an agent-assisted revision trail instead of untraceable rewriting
Prepare a rebuttal or a presentation storyboard from the same evidence base
The same three-skill bundle applied to rebuttal preparation and presentation storyboarding
Authors replying to reviewers or turning a paper into slides
02How to install into DeepSeek Harness
Prerequisites
- Python 3.10+ and Node.js for the skills-CLI installation path
- For the plugin path: DSH 0.2.1 or newer — the bundle needs no build step
Installation steps
- 01
Plugin path: `dsh plugin --profile web add dsh-paper-agent-skills`
$ dsh plugin --profile web add dsh-paper-agent-skills
- 02
Skills path: `npx skills add Scodive/Paper-Agent-Skills --agent universal --skill '*' --copy --yes` into the shared agent directory DSH can discover
$ npx skills add Scodive/Paper-Agent-Skills --agent universal --skill '*' --copy --yes
- 03
Then create the virtualenv and install the script requirements: `python3 -m venv .paper-agent-venv` and `.paper-agent-venv/bin/python -m pip install -r .agents/skills/academic-paper-writing/scripts/requirements.txt`
Verify the integration
- Open an agent session and ask it to revise a manuscript against your records; read the run output in `runs/review-01/results/`
- The README reports live DSH skill loading 3/3 on DSH 0.2.2-rc.1
03DSH integration and capability boundaries
DSH plugin bundle (`dsh-paper-agent-skills`) registering the three skills as a bundled skill provider, or the same skills installed into the shared agent directory with the skills CLI
Academic paper revision
Markdown, single-file TeX or text-based PDF; experiment records in JSON/CSV; references in BibTeX/JSON→suggestions.md, revised.md / revised.tex, changes.diff and checks.md in a run directory
the host agent performs the review; the bundled Python tools only package evidence and validate editsEvidence packaging and edit validation
supplied experiment records and references→packaged evidence plus mechanical validation of the edits
requires no model API keyRebuttal preparation and presentation storyboarding
the same paper and evidence base→rebuttal drafts and slide storyboards from the same skill bundle
04Who is it for? When not to use it?
Good for
- Researchers and students who want an agent-assisted revision trail instead of untraceable rewriting
- Authors replying to reviewers or turning a paper into slides
Not for
- The published validation results do not establish writing-quality gains, scientific correctness or acceptance-rate improvements.
- Experimental records are not reproduced, references are not independently verified, and PDF layout is not preserved by the tools; collected papers keep their own licenses.
05Compatibility, maintenance and safety notes
- The published validation results do not establish writing-quality gains, scientific correctness or acceptance-rate improvements.
- Experimental records are not reproduced, references are not independently verified, and PDF layout is not preserved by the tools; collected papers keep their own licenses.
- The skills-CLI path needs Python 3.10+ and Node.js; the DSH plugin path needs DSH 0.2.1 or newer and no build step.
- The bundled Python tools require no model API key, but the host agent performing the review still needs a working model in DSH.
MIT · README evidence only, no tagged release in the facts pack, safety not audited
06Frequently asked questions
How do I install it as a DSH plugin?
Run `dsh plugin --profile web add dsh-paper-agent-skills`. The bundle registers the three skills as a bundled skill provider in the model's skill catalog, requires DSH 0.2.1 or newer and needs no build step.
What is the alternative installation path?
The same three skills install through the skills CLI into the shared agent directory DSH can discover: `npx skills add Scodive/Paper-Agent-Skills --agent universal --skill '*' --copy --yes`, then create `.paper-agent-venv` and install the script requirements. That path needs Python 3.10+ and Node.js.
Do the tools call a model?
No — the bundled Python tools require no model API key. The host agent performs the review; the tools package the evidence and validate the edits mechanically.
What does a run produce?
A run directory such as `runs/review-01/results/` with `suggestions.md` (recommendations and unresolved items), `revised.md` or `revised.tex`, `changes.diff` (the exact edits) and `checks.md` (mechanical checks and evidence limitations).
Do the validation results prove better papers?
No. The README states plainly that the observed results do not establish writing-quality gains, scientific correctness or acceptance-rate improvements, and that references are not independently verified and PDF layout is not preserved.
07Related DSH workflows
nuwa-skill
by alchaincyf
Distills a named person or topic into a reusable perspective skill for agent sessions.
distilly
by titanwings
Meta-skill turning source material into a versioned Person Profile skill with incremental merge, correction layer, rollback, per-host install.
dsh-market
by dsh-market
插件市场 (plugin marketplace): 打开 Settings → Plugin Market 浏览/搜索/一键安装社区插件；安装源限制为 curated awesome-dsh-plugin 注册表（其余拒绝），默认禁用构建脚本（pnpm>=10 需显式开启）。
answer-me-with-html
by qingyuna
Turns multi-step questions into readable one-page HTML explainers with diagrams and optional narrated videos.
08Data and sources
The same three skills also ship as a DSH plugin bundle, which registers them as a bundled skill provider in the model's…
Either installation path works. The plugin bundle requires DSH 0.2.1 or newer and needs no build step.
Live DSH skill loading | 3/3 skills loaded | DSH 0.2.2-rc.1; skills installed by skills CLI 1.7.6
This page is generated from the project’s public documentation, repository metadata and a structured parse of DSH Plugins; last verified on 2026-11-11. Found an error? Submit a correction.
Best DeepSeek Harness Plugins
Twelve plugins worth installing first — picked from the whole catalog, across every category.