agentsclimarketplace

Research paper plan

Skill kenantang/codex-and-claude-skills/collected-academic-research-skills/sources/yananlong__codex-skills/research/research-paper-plan

Turn supported claims and evidence into a research paper plan that works either as a standalone manuscript-structuring pass or as the paper stage inside a coordinated research workflow. Use when asked to outline a paper, map claims to evidence, plan figures and citations, align experiments to manuscript claims, or keep limitations and threats to validity explicit.From its SKILL.md

Install
npx -y skills add kenantang/codex-and-claude-skills --skill research-paper-plan

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

4.6 KB, 950 tokens by cl100k_base, as published. Nobody here has run it

Research Paper Plan

Quick start

  1. Start from supported claims, not from section titles.
  2. Decide whether this is a standalone paper plan or the paper stage inside an orchestrated suite.
  3. Initialize paper-plan.md, claims-evidence-matrix.md, figure-plan.md, and citation-plan.md only when tracked outputs will help.
  4. Block on missing evidence instead of writing around it.
  5. Plan figures, citations, and limitations as first-class parts of the manuscript story.

Modes

Standalone mode

  • Work from the user prompt plus any local claims, results, notes, or drafts already present.
  • Do not require a suite root.
  • Collaboration is still allowed: if novelty review, experiment plan, results audit, or review-loop artifacts exist, use them; if another skill would materially improve the manuscript plan, recommend or invoke it.

Orchestrated mode

  • Prefer the canonical directory ./paper-plan/.
  • Read upstream context from research-brief.md, artifact-index.md, ./novelty-review/, ./experiment-plan/, ./review-loop/, and ./zotero/ when present.
  • Keep the paper outputs legible to later drafting and review passes.

Input contract

  • Minimum:
    • one or more claims the paper wants the reader to accept
    • evidence or source artifacts that support those claims
  • Prefer:
    • target venue or audience
    • known weaknesses or reviewer objections
    • desired exhibit list
    • existing novelty, experiment, results-audit, or review-loop artifacts
    • existing Zotero exports or library artifacts

Hard stops

  • Stop if the user wants a paper plan but there is no credible evidence behind the main claim.
  • Stop if claims outrun the available artifacts and the gap cannot be made explicit.
  • Stop if the venue constraints matter but remain unknown.
  • In standalone mode, do not create excess structure when a compact direct outline is enough.

Output contract

  • Primary files:
    • paper-plan.md
    • claims-evidence-matrix.md
    • figure-plan.md
    • citation-plan.md
  • In orchestrated mode, these live under ./paper-plan/.
  • In standalone mode, any target directory is valid.

Workflow

1) Freeze the supported claims

  • Use references/claims-evidence-matrix-template.md.
  • Separate:
    • main contribution claims
    • supporting claims
    • limitation claims
  • Downgrade or remove claims that do not have a clear evidence path.

2) Fit the paper shape to the evidence

  • Use references/paper-outline-template.md, references/section-archetypes.md, and references/venue-adapters.md.
  • Choose a structure that matches what the evidence can actually support.
  • Treat page budget and venue expectations as constraints, not afterthoughts.

3) Plan exhibits deliberately

  • Use references/exhibit-plan-checklist.md.
  • Every figure and table should answer a paper question or defend a claim.
  • Mark exhibits as:
    • mandatory
    • helpful
    • cut

4) Build the citation plan carefully

  • Use references/citation-verification-rules.md.
  • If a Zotero library already exists for the project, invoke research-zotero or consume ./zotero/ artifacts before inventing citation structure from memory.
  • Record which claims need citations versus which require empirical evidence.
  • Keep citation verification explicit instead of inventing references from memory.

5) Keep collaboration open

  • Pull in research-results-auditor when the evidence quality is unclear.
  • Pull in research-review-loop when the story needs an adversarial pass.
  • Pull in research-novelty-review when positioning remains unstable.
  • Pull in research-zotero when citation grounding, BibTeX export, or CSL-JSON export would materially improve the paper plan.
  • Missing evidence remains a block in every mode. Do not write unsupported paper stories.

References

  • references/paper-outline-template.md
  • references/claims-evidence-matrix-template.md
  • references/section-archetypes.md
  • references/venue-adapters.md
  • references/exhibit-plan-checklist.md
  • references/citation-verification-rules.md
  • references/tabmol-ddi-ood-adapter.md

Script

  • scripts/init_paper_pack.py: create the core paper-planning files in a standalone directory or the suite's paper-plan/ directory.

What ships with it: 9 files

5.4 KB alongside SKILL.md, 1 of them executable

agents/

scripts/

Keep looking

Skills are one crate of 326,750. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.