Research Writing
alfonso0512/research-writing-skill
科研论文写作助手，提供 30 个 Prompt 模板覆盖论文写作全流程. An agent skill from alfonso0512/research-writing-skill.
Build or resume mathematical-modeling competition work, including CUMCM, MCM/ICM, graduate and regional contests, data challenges, and open-topic statistical modeling.
Install Cumcm Workflow
$ npx skills add Lucasuiii/modeling-workbench --skill cumcm-workflow -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install Lucasuiii/modeling-workbench cumcm-workflow --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/Lucasuiii/modeling-workbench.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/cumcm-workflow .claude/skills/cumcm-workflow && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "cumcm-workflow" agent skill from https://github.com/Lucasuiii/modeling-workbench/tree/main/.agents/skills/cumcm-workflow into .claude/skills/cumcm-workflow/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cumcm-workflow", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/Lucasuiii/modeling-workbench/tree/main/.agents/skills/cumcm-workflowType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add Lucasuiii/modeling-workbench --skill cumcm-workflow -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install Lucasuiii/modeling-workbench cumcm-workflow --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Lucasuiii/modeling-workbench.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/cumcm-workflow .agents/skills/cumcm-workflow && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "cumcm-workflow" agent skill from https://github.com/Lucasuiii/modeling-workbench/tree/main/.agents/skills/cumcm-workflow into .agents/skills/cumcm-workflow/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cumcm-workflow", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add Lucasuiii/modeling-workbench --skill cumcm-workflow -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install Lucasuiii/modeling-workbench cumcm-workflow --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Lucasuiii/modeling-workbench.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/cumcm-workflow .cursor/skills/cumcm-workflow && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "cumcm-workflow" agent skill from https://github.com/Lucasuiii/modeling-workbench/tree/main/.agents/skills/cumcm-workflow into .cursor/skills/cumcm-workflow/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cumcm-workflow", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/Lucasuiii/modeling-workbench.git --path .agents/skills/cumcm-workflow--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add Lucasuiii/modeling-workbench --skill cumcm-workflow -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install Lucasuiii/modeling-workbench cumcm-workflow --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Lucasuiii/modeling-workbench.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/cumcm-workflow .gemini/skills/cumcm-workflow && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "cumcm-workflow" agent skill from https://github.com/Lucasuiii/modeling-workbench/tree/main/.agents/skills/cumcm-workflow into .gemini/skills/cumcm-workflow/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cumcm-workflow", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install Lucasuiii/modeling-workbench cumcm-workflowInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add Lucasuiii/modeling-workbench --skill cumcm-workflow -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/Lucasuiii/modeling-workbench.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/cumcm-workflow .github/skills/cumcm-workflow && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "cumcm-workflow" agent skill from https://github.com/Lucasuiii/modeling-workbench/tree/main/.agents/skills/cumcm-workflow into .github/skills/cumcm-workflow/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cumcm-workflow", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add Lucasuiii/modeling-workbench --skill cumcm-workflow -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install Lucasuiii/modeling-workbench cumcm-workflow --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/Lucasuiii/modeling-workbench.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/cumcm-workflow .opencode/skills/cumcm-workflow && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "cumcm-workflow" agent skill from https://github.com/Lucasuiii/modeling-workbench/tree/main/.agents/skills/cumcm-workflow into .opencode/skills/cumcm-workflow/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "cumcm-workflow", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
cumcm-workflow- GitHub stars
- 172
- Token cost
- ~2.3k tokens
- SKILL.md length
- 1,051 words
- Files
- 93 (incl. scripts, references, assets)
- Skills in repo
- 2
- Repo updated
- First seen
- Licence
- MIT
At a glance
Build or resume mathematical-modeling competition work, including CUMCM, MCM/ICM, graduate and regional contests, data challenges, and open-topic statistical modeling.
- Works in 5 steps: Read the project's .cumcm/state.json and… → For a new project, use intake and… → Read only the active stage guide below.… → …
- Tasks that involve LaTeX
- SKILL.md covers Competition and task routing, Start or resume, Setup and progress diagnostics and Three human stops, plus 3 more sections
- Calls python3
What it does
Cumcm Workflow is an agent skill from Lucasuiii/modeling-workbench. Build or resume mathematical-modeling competition work, including CUMCM, MCM/ICM, graduate and regional contests, data challenges, and open-topic statistical modeling. Guide problem framing, modeling, computation, validation, Chinese or English LaTeX writing, and reviewed PDF/source delivery using current official requirements. Not for ordinary paper polishing.
Its SKILL.md is about 2.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 98 other files, including scripts, reference files and assets (for example `agents/openai.yaml` and `assets/latex-template/generic-ctex/template.json`).
It sits in Documents & Office, covering LaTeX. It works with LaTeX. The repository describes itself as: Contest-native, evidence-focused math modeling workflow for Codex and Claude Code. The licence is MIT.
When your agent uses it
- Tasks that involve LaTeX
Example prompts
- “/cumcm-workflow”
Requirements
- Python 3
Workflow steps
5 steps, taken from the first numbered list in SKILL.md.
- Read the project's .cumcm/state.json and the incoming handoff, if present. Resume exact 0.6.0 projects; older schemas are unsupported.
- For a new project, use intake and init_project.py with the supplied official files.
- Read only the active stage guide below. Consult a schema only when the guide and command help leave a specific field unresolved; do not…
- Define S as the absolute path to this Skill's scripts directory. All examples use python3 "$S/.py"; the contest workspace does not contain…
- After interruption, check pending human decisions before continuing. Existing downstream files do not establish approval. Never infer…
What it can do on your machine
Read from SKILL.md and the folder at commit b4af5f7. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Ships 1 file in scripts/, which the agent can run.
Shell commands in SKILL.md call:
python3
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Cumcm Workflow loads about 2.3k tokens when it runs, and up to ~33k if it reads all its reference files. Until then it costs about 95 tokens; SKILL.md has 1,051 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
SKILL.md
The full file from Lucasuiii/modeling-workbench at commit b4af5f7, republished under its MIT licence (© Lucasuiii). 1,051 words, ~2,305 tokens.
Download SKILL.mdSave it as
.claude/skills/cumcm-workflow/SKILL.md (or your agent's skills folder). This skill also uses 92 other files; get the full folder from GitHub.- name
- cumcm-workflow
- description
- Build or resume mathematical-modeling competition work, including CUMCM, MCM/ICM, graduate and regional contests, data challenges, and open-topic statistical modeling. Guide problem framing, modeling, computation, validation, Chinese or English LaTeX writing, and reviewed PDF/source delivery using current official requirements. Not for ordinary paper polishing.
Modeling Workbench
Spend reasoning on the problem, mathematics, experiments and explanation. Tools maintain execution records, hashes, snapshots and stage state. Do not create extra checklists or repeatedly edit contracts to silence warnings.
Competition and task routing
For a new competition, read competition adaptation to identify the current official requirements and the supported automation boundary. Competition names do not determine methods, page limits or evidence standards. Keep the existing stages, two knobs and three human stops.
Read additional guidance only when the active work needs it:
| Current need | Read |
|---|---|
| Choose a research question under an official theme; find suitable data | Open-topic research during problem analysis |
| Choose a model, audit data, design a useful comparison within the available budget | Task-driven modeling during analysis/model design/computation |
| English summary, audience-specific memo, references or format adaptation | Competition writing during paper planning |
For paper initialization, pass the actual --competition and --language zh|en to init_latex_paper.py; omitted options retain CUMCM/Chinese behavior. The shared LaTeX/PDF compile, review and source-package chain works across competition names. Generic scaffolds are not official templates: current-rule compliance, page QA and the three human stops still apply. DOCX export is not implemented.
Start or resume
- Read the project's
.cumcm/state.jsonand the incoming handoff, if present. Resume exact0.6.0projects; older schemas are unsupported. - For a new project, use intake and
init_project.pywith the supplied official files. - Read only the active stage guide below. Consult a schema only when the guide and command help leave a specific field unresolved; do not load the checker to learn the workflow.
- Define
Sas the absolute path to this Skill'sscriptsdirectory. All examples usepython3 "$S/<command>.py"; the contest workspace does not contain these scripts. - After interruption, check pending human decisions before continuing. Existing downstream files do not establish approval. Never infer approval from a quota reset, a new task, or “continue”.
Setup and progress diagnostics
On first setup or an environment change, run python3 "$S/doctor.py"; on resumption or a progress question, run python3 "$S/project_status.py" --project <p>. Both print reports without changing project state. Summarize what is usable, what blocks the next action, and the next step; continue authorized work instead of adding a confirmation point. Missing paper tools do not block modeling. A ready preflight is not approval: read the separately reported checkpoint availability. Use diagnostics for probe depth, JSON output, optional model dependencies and exit codes. Do not run diagnostics on every reply or reinstall an already prepared environment.
Three human stops
| Before | Show the user | Record after their explicit reply |
|---|---|---|
| Official computation | Objective, constraints, all candidates and their discriminating evidence, chosen scope, any unanswered requirement | model-design |
| Paper writing | Every claim's text, scope, evidence state, and open P0/P1 | validation |
| Final delivery | Current PDF pages, answers, remaining findings and actual delivery files | delivery |
Stop the dependent work after presenting the material. Model self-review is useful judgement, never human acceptance. A reply before the material was shown does not approve it. If a reviewed claim or model changes, show the revision and obtain a new decision; do not relabel the old acceptance.
After the user accepts all the presented current material, one command fills the existing checkpoint, records its snapshot and advances state:
python3 "$S/record_decision.py" --project <p> --stage <stage> \
--decision accepted --confirm-human --task-turn-ref <user-reply-ref> \
--summary <what-the-user-accepted>No manual timestamps, presented-ID lists, hashes or state edits. Other stages are technical completions: use the same command without --confirm-human, after their checks pass, referencing the current task. They do not require another user confirmation. Reopen with --decision revision_requested; downstream approvals become unusable. Decisions are honest conversation records, not cryptographic proof that a person answered.
Show full SKILL.md (468 more words)Show less
Working and finalizing
workingpermits incomplete model drafts and cheap exploratory runs. Failed exploration never blocks. Usepreflightwhile drafting: pending review is visible but does not fail the command.enforcerequires the three human stops in both modes. Official recording and paper entry also check the corresponding stop, so skipping a checker does not silently replace approval with self-review.finalizingrequires complete current evidence, decisions, independent review and delivery binding. Switch withset_mode.py. Do not run full finalizing checks before exploratory model selection: formal assertions do not exist yet.
python3 "$S/cumcm_check.py" --project <p> --stage <stage> --gate-mode preflightWarnings remain visible; they are not a request to rewrite upstream evidence. An error requires repair; awaiting_review means present the material and wait. Passing does not prove mathematical correctness.
Stage guides
| Work | Read | Outgoing handoff |
|---|---|---|
| Problem analysis | 02-problem-analysis.md | — |
| Model candidates and cheap comparisons | 03-model-design.md | modeling-computation |
| One backend, official runs and result indexing | 04-computation.md | computation-validation |
| Independent review and conclusions | 05-validation.md | validation-paper |
| Reader-facing paper and visual QA | 06-paper-writing.md | paper-delivery |
| Actual delivery packages | 07-compile-delivery.md | final package |
Build handoffs with build_handoff.py; read handoffs only at a crossing. computation-validation and validation-paper must cross into fresh tasks. Task refs are a paste guard, not proof of independence; same-model new-context review remains correlated. After a full review finds P0, the package builder defaults to targeted re-review of those findings.
Evidence without paperwork
- Preserve official files.
record_run.pyrecords real execution and freezes declared evidence;index_result.pyreads values from outputs. Never type machine facts into contracts. - Choose a model after cheap candidate evaluation; officially implement one backend. MATLAB preference breaks ties, not task suitability. No parity implementation unless requested.
- Runs are append-only. A rerun uses
--rerun, never overwrites its parent. Only successful official descendants supersede.plan_redo.pyscopes affected work; it does not waive checks. - P0: wrong computation/data, task mismatch, scope beyond evidence, stale evidence, fabricated approval/review, or unusable delivery. P1: weaknesses within a supported and task-relevant scope. P2: optional improvements. Narrowing a claim cannot erase an unanswered requirement.
- Derive review priorities from the current task, not previous failure examples. Check task coverage, model assumptions, solution validity and claim scope using a few tests or independent arguments that could expose a plausible wrong answer. Choose applicable mathematical properties; do not require every problem to run the same tests. A repair must address the failure mechanism and affected conclusions, not only the failing example.
- Claims use
supported_not_reproducedunless an isolated rerun and comparison establishreproduced. Label simulations and synthetic scenarios explicitly. - Keep workflow IDs and evidence bookkeeping out of reader-facing prose. Compilation logs do not establish visual quality; inspect rendered pages.
refresh_evidence.py --only delivery --packagebuilds declared ZIPs with project-relative directories and refreshes their existing metadata. It never refreshes official sources. No-change refreshes do not rewrite files.
Use artifact contracts only for an unfamiliar artifact and evidence rules for unresolved evidence semantics. Do not read every reference at startup.
© Lucasuiii, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 92 other files (scripts, references, assets) in .agents/skills/cumcm-workflow of Lucasuiii/modeling-workbench.
- SKILL.md
- agents/openai.yaml
- assets/latex-template/generic-ctex/macros.tex
- assets/latex-template/generic-ctex/main.tex.tmpl
- assets/latex-template/generic-ctex/metadata.tex.tmpl
- assets/latex-template/generic-ctex/planned-section.tex.tmpl
- assets/latex-template/generic-ctex/references.bib
- assets/latex-template/generic-ctex/sections/00_abstract.tex
- assets/latex-template/generic-ctex/sections/98_references.tex
- assets/latex-template/generic-ctex/sections/99_appendix.tex
- assets/latex-template/generic-ctex/template.json
- assets/latex-template/generic-en/macros.tex
- assets/latex-template/generic-en/main.tex.tmpl
- assets/latex-template/generic-en/metadata.tex.tmpl
- assets/latex-template/generic-en/planned-section.tex.tmpl
- … and 78 more
Open the folder on GitHubat commit b4af5f7
Compare with similar skills
Cumcm Workflow next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Cumcm Workflow this skillLucasuiii/modeling-workbench | 172 | — | ~2.3k | Automated safety check: Pass | MIT | |
| Research Writingalfonso0512/research-writing-skill | 488 | 1 repos | ~818 | Automated safety check: Pass | MIT | |
| Paper WritingMLNLP-World/Paper-Writing-Tips | 4.7k | — | ~630 | Automated safety check: Pass | None | |
| Evomath TaoEvoScientist/EvoSkills | 476 | 2 repos | ~3.8k | Automated safety check: Pass | Apache-2.0 | |
| PaperjurySpark-To-Paper-Skills/paperjury | 1.2k | — | ~5.3k | Automated safety check: Pass | MIT | |
| Thesis Defense PPTX Builderzouchenzhen/thesis-defense-pptx-skill | 266 | — | ~2.4k | Automated safety check: Pass | Apache-2.0 |
Similar skills
Paper Writing
MLNLP-World/Paper-Writing-Tips
学术论文写作检查与优化助手。基于 MLNLP-World 社区整理的论文写作技巧，帮助检查和优化学术论文。Use when: (1) 检查论文 LaTeX 格式和排版, (2) 优化公式符号使用, (3) 改进图表设计, (4) 润色英文学术表达, (5) 检查参考文献格式, (6) 投稿前终稿检查, (7) 用户询问论文写作技巧或规范。
Evomath Tao
EvoScientist/EvoSkills
A skill your agent uses whenever the user submits a non-trivial mathematical claim that needs a rigorous proof or audit.
Paperjury
Spark-To-Paper-Skills/paperjury
Three modes for CS-conference papers (CVPR/ICCV/ECCV vision, ACL/EMNLP/NAACL NLP, ICLR/NeurIPS/ICML/AAAI ML).
Thesis Defense PPTX Builder
zouchenzhen/thesis-defense-pptx-skill
Builds an editable thesis defense PowerPoint from a thesis PDF or LaTeX project while preserving a supplied university or lab template, then runs a visual quality check.
Mathmodel Skill
handsomeZR-netizen/mathmodel-skill
CUMCM 国赛、MCM/ICM 美赛与电工杯数学建模竞赛的端到端协作工作流。Use when a user explicitly works on one of these modeling contests or asks to run/review a modeling-competition paper from problem selection through modeling…
More from Lucasuiii/modeling-workbench
Cumcm Independent Review
Lucasuiii/modeling-workbench
Independently review a CUMCM computation package before its claims enter validation or paper writing.
Related
Works with
Categories
Questions about Cumcm Workflow
What does Cumcm Workflow do?
Build or resume mathematical-modeling competition work, including CUMCM, MCM/ICM, graduate and regional contests, data challenges, and open-topic statistical modeling. Cumcm Workflow is an agent skill from Lucasuiii/modeling-workbench. Build or resume mathematical-modeling competition work, including CUMCM, MCM/ICM, graduate and regional contests, data challenges, and open-topic statistical modeling.
When should I use Cumcm Workflow?
Cumcm Workflow fits situations like: tasks that involve LaTeX.
How do I install Cumcm Workflow in Claude Code?
Run `npx skills add Lucasuiii/modeling-workbench --skill cumcm-workflow -a claude-code`. Or copy the skill folder (.agents/skills/cumcm-workflow in Lucasuiii/modeling-workbench) into .claude/skills/cumcm-workflow in your project. Claude Code loads it when a task matches its description.
How do I install Cumcm Workflow in Codex?
Run `npx skills add Lucasuiii/modeling-workbench --skill cumcm-workflow -a codex`. Or copy the skill folder (.agents/skills/cumcm-workflow in Lucasuiii/modeling-workbench) into .agents/skills/cumcm-workflow in your project. Codex loads it when a task matches its description.
Can I use Cumcm Workflow in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Lucasuiii/modeling-workbench --skill cumcm-workflow -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/cumcm-workflow, .gemini/skills/cumcm-workflow, .github/skills/cumcm-workflow and .opencode/skills/cumcm-workflow in your project.
What does Cumcm Workflow need to run?
Going by SKILL.md and its folder, Cumcm Workflow needs the command-line tools its instructions call (python3). Our summary lists: Python 3.
Does Cumcm Workflow access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Cumcm Workflow safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
What licence does Cumcm Workflow use?
Cumcm Workflow is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Cumcm Workflow use?
About 2.3k tokens (SKILL.md is roughly 9.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 30k tokens, read only when the agent opens those files.
What are the alternatives to Cumcm Workflow?
Skills that share tags, products or a category with Cumcm Workflow: Research Writing (alfonso0512/research-writing-skill, 488 stars), Paper Writing (MLNLP-World/Paper-Writing-Tips, 4.7k stars), Evomath Tao (EvoScientist/EvoSkills, 476 stars) and Paperjury (Spark-To-Paper-Skills/paperjury, 1.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Cumcm Workflow?
Lucasuiii (a GitHub user) maintains it in Lucasuiii/modeling-workbench, which has 172 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on September 23, 2026.
Source: Lucasuiii/modeling-workbench on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.