A research-writing tool that turns a connected set of validated research ideas and evidence into a structured paper outline. It plans sections, figures, and citations for venues such as ICLR, NeurIPS, and IEEE.

Use it to plan a machine-learning or computer-science paper from existing research notes, including its narrative, section order, figures, and citations.

It removes the need to manually connect experiments, claims, and paper sections. A review step also checks whether the planned story is convincing.

How skyllwt/AutoSci configures Claude Code

About the project

AutoSci is an AI research platform organized around a wiki, with an agent that supports stages of scientific work such as reading, experimentation, writing, and retaining knowledge across projects. It is for people building or using AI-assisted research workflows, with Claude Code, Codex, and OpenCode adaptations available. The catalogue add-ons extend those agent-specific workflows.

skyllwt/AutoSci · 1,663 stars · on GitHub

Reuse

Borrowing it

Nothing to install: this file belongs to skyllwt/AutoSci. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/skyllwt/AutoSci/main/.claude/skills/paper-plan/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/skyllwt/AutoSci

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/skyllwt/autosci/paper-plan.svg)](https://agentmods.dev/skills/skyllwt/autosci/paper-plan)

Your own site

<a href="https://agentmods.dev/skills/skyllwt/autosci/paper-plan"><img src="https://agentmods.dev/badge/skills/skyllwt/autosci/paper-plan.svg" alt="Measured on agentmods" height="20"></a>

Per session 29 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 4,171 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

What 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 5.1	$0.00029	$0.04171
Opus 5	$0.00015	$0.02086
Sonnet 5	$0.00006	$0.00834
Haiku 4.5	$0.00003	$0.00417

Measured 9d ago against content hash ffeb06a89767, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

paper-plan 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 9d 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.

.claude/skills/paper-plan/SKILL.md · 400 lines

How it starts

The opening of the file, as written. The whole thing — 400 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/paper-plan

Compile a paper outline from the wiki's idea graph. Input target ideas (status: validated or in-progress with succeeded experiments), specify the target venue, compile an evidence map from the wiki → determine narrative structure → generate a section outline + figure plan + citation plan. Review LLM review is a mandatory step (acting as area chair to assess outline persuasiveness). Output PAPER_PLAN.md to wiki/outputs/.

Key distinction: the outline is idea-graph-driven — each section exists because it supports an idea (or its evidence/methods), not because paper convention requires that section.

Inputs

ideas: list of target idea slugs (space-separated)
- each idea should have status: validated or be in_progress with at least one succeeded experiment
- if proposed or invalidated ideas are included, warn but continue
--venue (required): target venue, determines page limit and format requirements
- supported: ICLR / NeurIPS / ICML / ACL / CVPR / IEEE
--title (optional): working title; if omitted, generated from target ideas

Outputs

wiki/outputs/paper-plan-{slug}-{date}.md — complete paper plan (PAPER_PLAN.md)
wiki/graph/edges.jsonl — new derived_from edges (plan → source ideas/papers)
wiki/graph/context_brief.md — rebuilt
wiki/log.md — appended log entry
PAPER_PLAN_REPORT (printed to terminal) — plan summary

Wiki Interaction

Reads

wiki/ideas/*.md — Hypothesis, Motivation, Approach sketch, Novelty argument, status, novelty_score, target_venue, linked_experiments, origin_gaps
wiki/experiments/*.md — supporting experiments (linked via linked_idea); results, metrics, key_result
wiki/methods/*.md — methods referenced by the idea's Approach sketch (Mechanism, Procedure, source_papers)
wiki/papers/*.md — evidence source papers (Method, Results, Related)
wiki/concepts/*.md — concepts the idea's origin_gaps points to (Definition, Variants, Comparison)
wiki/topics/*.md — topics the idea's origin_gaps points to (Overview, Open problems)
wiki/graph/context_brief.md — global context
wiki/graph/open_questions.md — knowledge gaps (annotate paper limitations)
wiki/graph/edges.jsonl — relationship graph (build narrative logic chain)
.claude/skills/shared-references/academic-writing.md — writing principles
.claude/skills/shared-references/citation-verification.md — citation discipline

Read the full file on GitHub · 400 lines

Changes

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.

9d ago First seen · 400 lines · 29 tokens per session scan A ffeb06a89767

Subscribe to this mod's changes

paper-plan is a skill published in the GitHub repository skyllwt/AutoSci (1,663 stars, last pushed 2d ago), licensed MIT. It adds 29 tokens to every session and 4,171 once invoked, about $0.0001 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.

Research Writing and docs

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens

chronicle

Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…

microsoft/vscode · 72 tokens

AutoSci: Skill for Claude Code