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/xiaomimimo/mimo-code/research-paper-writingnpx skills add XiaomiMiMo/MiMo-Code --skill research-paper-writinggit clone --depth 1 https://github.com/XiaomiMiMo/MiMo-CodeWhat 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 | $0.00161 | $0.01561 |
| Opus 5 | $0.00081 | $0.00781 |
| Sonnet 5 | $0.00032 | $0.00312 |
| Haiku 4.5 | $0.00016 | $0.00156 |
Grade A, and why
research-paper-writing 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 3d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Research Paper Writing
Act as an experienced co-author who has published at top ML/CV/NLP venues. The goal is not "prettier English" — it is a draft that survives skeptical reviewers: clear story, one message per paragraph, and every claim backed by evidence.
Step 1: Diagnose the request, then route
Match the user's situation and load ONLY the needed reference (do not preload all):
| User situation | What to do | Load |
|---|---|---|
| "Polish this paragraph / does this flow?" | Quick polish pass (see below) | references/paragraph-flow.md |
| Draft or rewrite Abstract | Pick a template, draft, check claims | references/abstract.md |
| Draft or rewrite Introduction | Clarify story first, then template | references/introduction.md |
| Draft or rewrite Related Work | Topic grouping + positioning | references/related-work.md |
| Draft or rewrite Method | Module triad workflow | references/method.md |
| Draft or rewrite Experiments; fix tables | Claim→experiment mapping, table rules | references/experiments.md |
| Draft or rewrite Conclusion / Limitations | Scope-based limitation framing | references/conclusion.md |
| "Review my paper / 投稿前检查" | Adversarial five-dimension review | references/paper-review.md |
| "Compile / convert to PDF / 编译不过" | Pick engine, build, verify output | references/pdf-export.md |
| Bullet points / Chinese notes → English section | Treat as section drafting; get facts first, then draft | the matching section guide |
If the user's draft lives in files (.tex, .md, Overleaf export), work on the files directly: locate sections with grep (\section, \begin{abstract}), edit in place, and keep every \cite{...}, \ref{...}, label, comment, and math environment byte-for-byte unless the edit is specifically about them.
Step 2: Get the facts before writing
Good paper writing is mostly correct facts arranged well. Before drafting a section, make sure these are known — from the draft, the codebase, or the user:
- What exact technical problem is solved, and why prior methods fail at it (limitation + technical reason).
- What the contribution is (new task / pipeline / module / finding / insight).
- Why the method works in essence, and its concrete advantages.
- What the strongest experimental numbers are.
What ships with it
35 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.
- agents/openai.yaml 363 B
- references/abstract.md 3.3 KB
- references/conclusion.md 1.2 KB
- references/examples/abstract/template-a.md 950 B
- references/examples/abstract/template-b.md 3.2 KB
- references/examples/abstract/template-c.md 1.7 KB
- references/examples/introduction/novel-task-challenge-decomposition.md 1.5 KB
- references/examples/introduction/pipeline-not-recommended-abstract-only.md 2.9 KB
- references/examples/introduction/pipeline-version-1-one-contribution-multi-advantages.md 2.0 KB
- references/examples/introduction/pipeline-version-2-two-contributions.md 2.0 KB
- references/examples/introduction/pipeline-version-3-new-module-on-existing-pipeline.md 1.5 KB
- references/examples/introduction/pipeline-version-4-observation-driven.md 1.4 KB
- references/examples/introduction/technical-challenge-version-1-existing-task.md 2.6 KB
- references/examples/introduction/technical-challenge-version-2-existing-task-insight-backed-by-traditional.md 3.8 KB
- references/examples/introduction/technical-challenge-version-3-novel-task.md 1.6 KB
- references/examples/introduction/version-1-task-then-application.md 727 B
- references/examples/introduction/version-2-application-first.md 405 B
- references/examples/introduction/version-3-general-to-specific-setting.md 847 B
- references/examples/introduction/version-4-open-with-challenge.md 1.7 KB
- references/examples/method/example-of-the-three-elements.md 7.3 KB
- references/examples/method/method-writing-common-issues-note.md 355 B
- references/examples/method/module-design-instant-ngp.md 4.5 KB
- references/examples/method/module-motivation-patterns.md 410 B
- references/examples/method/module-triad-neural-body.md 763 B
- references/examples/method/neural-body-annotated-figure-text.md 2.8 KB
- references/examples/method/overview-template.md 2.0 KB
- references/examples/method/pre-writing-questions.md 479 B
- references/examples/method/section-skeleton.md 110 B
- references/experiments.md 4.4 KB
- references/introduction.md 14 KB
- references/method.md 6.0 KB
- references/paper-review.md 5.3 KB
- references/paragraph-flow.md 2.1 KB
- references/pdf-export.md 4.0 KB
- references/related-work.md 1.3 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.
- 3d ago First seen · 89 lines · 161 tokens per session scan A 296435acea1f
research-paper-writing is a skill published in the GitHub repository XiaomiMiMo/MiMo-Code (12,923 stars, last pushed yesterday), licensed MIT. It adds 161 tokens to every session and 1,561 once invoked, about $0.0008 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
instrument-data-to-allotrope
Convert laboratory instrument output files (PDF, CSV, Excel, TXT) to Allotrope Simple Model (ASM) JSON format or flattened 2D CSV. Use this skill when scientists need to standardize instrument data for LIMS systems, data lakes, or downstream analysis. Supports auto-detection of instrument types. Outputs include full…
exploratory-data-analysis
Perform bounded, local exploratory analysis of explicitly supported scientific files. Use for redacted CSV/TSV/JSON profiles; optional NumPy, HDF5, FASTA/FASTQ, and basic image metadata inspection; missingness/leakage audits; outlier and transformation sensitivity; and rigorous EDA report scaffolds. Other domain…
mapping-to-snomed
Maps clinical concept spans extracted by OpenMed to SNOMED CT concepts through a USER-SUPPLIED terminology server (the user's own Ontoserver, Snowstorm, or UMLS/UTS), never a bundled vocabulary. Use when the user wants to code findings, disorders, procedures, body structures, or substances to SNOMED CT, run an ECL…
auditing-subgroup-fairness
Audit an OpenMed NER or de-identification model for performance disparities across demographic subgroups (sex, age band, race/ethnicity when available) using openmed.eval.fairnessreport. Use when the user wants per-subgroup recall and leakage, wants to check whether de-identification under-protects a group, wants to…
overleaf-sync
Two-way sync between a local paper directory and an Overleaf project, so ARIS audit/edit workflows stay on the local copy while collaborators edit in the Overleaf web UI. Use when user says "同步 overleaf", "overleaf sync", "推送到 overleaf", "connect overleaf", "Overleaf 桥接", "pull overleaf", "push overleaf", or wants to…
mixed-precision
Use FP16/BF16 mixed precision to accelerate training and reduce memory. Use when optimizing GPU performance.