Skip to content

sidiangongyuan/codex-skills-library / skills/paper-framework-figure-studio-pro

paper-framework-figure-studio-pro

Verified

Use when planning a source-grounded method overview, architecture, pipeline, system/data-flow, mechanism, or agent-workflow figure for a CS or deep-learning paper. Produces an editable figure brief before image generation or manual redrawing; not for plots, tables, or general prose.

FileKindSize
LICENSEother1.0 KB
SKILL.mdskill md5.0 KB
agents/openai.yamlconfig313 B
references/source-map.mdmarkdown2.3 KB
references/workflow.mdmarkdown2.9 KB
templates/figure-brief-template.mdmarkdown755 B

references/workflow.md

View on GitHub
# Local S0-S5 Workflow

Use this reference when the user asks for a staged framework-figure planning
session or when the figure is important enough to need a source-faithful prompt
contract.

## S0 Paper Foundation

Goal: understand what the paper actually supports before designing anything.

Collect:

- source files or text excerpts
- target contribution claim
- modules, data types, training/inference split, losses, variables, and symbols
- evidence anchors for arrows, dependencies, and evaluation claims
- known ambiguity, missing source, or unsupported visual risks

Output a compact risk register. If the core method is underspecified, stop and
ask for the missing source.

## S1 Figure Strategy

Goal: decide what the first glance should teach.

Define:

- figure role and reader question
- primary module hierarchy
- semantic graph: entities and relationships
- visual render graph: layout regions, ordering, grouping, and connectors
- visible text contract: exact labels allowed in the figure
- caption/prose handoff: details that should not be inside the image
- style family: publication schematic, clean line art, minimal flat diagram, or
  user-specified surface style

Output candidate directions or prompt packages, but do not generate images
unless explicitly requested.

## S2 Sketch Exploration

Goal: explore diverse layouts without changing the paper semantics.

Default local behavior is text-only: propose candidate directions such as
pipeline-left-to-right, hierarchical module stack, data/model/control split,
agent loop, or uncertainty/diagnostic panel.

Generate images only if the user explicitly asks. Smoke-test artifacts or toy
outputs created for validation must be deleted after the test unless the user
asked to keep them.

## S3 Direction Select

Goal: choose or combine directions based on evidence and user preference.

Review each candidate for:

- source fidelity
- first-glance clarity
- connector correctness
- symbol consistency
- text density
- whether visual details belong in caption/prose instead

Record what to preserve, remove, simplify, or repair.

## S4 Formal Candidate Brief

Goal: turn the selected direction into a handoff-ready figure specification.

Include:

- final candidate matrix
- exact module labels and connector labels
- must-include and must-exclude lists
- allowed symbols, colors, and line styles
- prompt package or manual-redraw brief
- residual risks and required user confirmations

## S5 Candidate Image

Goal: optionally generate final candidate images.

Only run this stage when the user explicitly asks for image generation and the
source-faithful contract is ready. After S5, stop. The human owns candidate
selection, manual editing, caption finalization, and paper placement.

## Compact Output Format

For most user requests, do not expose all internals. Return:

- figure role
- reader question
- source-backed core modules
- semantic flow
- visible text contract
- style recommendation
- candidate directions
- risks
- next action
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 322kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall