Paper section playbook
Skill sidiangongyuan/codex-skills-library/skills/paper-section-playbook
Use when planning, drafting, or restructuring the Abstract, Introduction, Related Work, Method, Experiments, or Conclusion of a computer-vision, 3D-perception, or autonomous-driving paper. Provides section- and paragraph-level methodology; use paper-refinement-skills for sentence-level polish.From its SKILL.md
npx -y skills add sidiangongyuan/codex-skills-library --skill paper-section-playbookAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 24 stars24 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.
What its file declares
Copied from the file, not written here
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
4.4 KB, 918 tokens by cl100k_base, as published. Nobody here has run it
Paper Section Playbook
Scope and complement
This skill answers "how should this section be structured and what should each paragraph do?" for papers in CV, 3D perception, and autonomous driving venues (CVPR / ICCV / ECCV / NeurIPS / CoRL).
For sentence-level prose polish (word choice, punctuation, terminology consistency, em-dash rules, etc.), defer to paper-refinement-skills instead. Both skills may be active at once; this one operates at the section and paragraph level, the other at the sentence level.
When to load this skill
- The user is planning or restructuring a paper section, not just polishing prose.
- The user asks "how do I write the Intro / Method / Experiments / etc."
- The user is rebutting a reviewer comment about section structure, motivation, or framing.
- The user wants exemplar-based guidance ("how do top papers do X").
Routing
- Identify which section the user is working on.
- Read the matching file under
sections/for the structural playbook (paragraph functions, decision tree, common reviewer red flags). - If the user is choosing between writing styles, consult
decision-trees/paper-type-router.mdto identify their paper type (new method / new task / new analysis / system / dataset), then apply the corresponding branch in the section file. - For Related Work, method motivation, experiment framing, or exemplar-based guidance, use
$research-evidencewhen citation support or top-venue precedent is uncertain. Prefer CVPR, ICCV, ECCV, ICLR, NeurIPS/NIPS, and ICML before secondary autonomous-driving or robotics venues unless the topic requires them. - Use the worked skeletons and templates in the relevant section file when the user wants concrete patterns.
- After drafting, run the section file's self-review checklist and
checklists/reviewer-redflags.mdfor the relevant section.
Authoring principle
Section files combine patterns from cited, high-impact exemplar papers with the
generalized principles in references/design-rationale.md. Treat exemplars as
evidence, not authority. Keep a rule only when it has a traceable source or a
clear rationale with stated conditions.
Do not invent generic writing advice. If a rule has no source, it does not belong here.
Section files
sections/abstract.md— authoredsections/introduction.md— authoredsections/related-work.md— authoredsections/method.md— authoredsections/experiments.md— authoredsections/conclusion.md— authored
(Files appear as the corresponding section is elicited and ratified by the user.)
Post-use feedback loop (meta-rule)
This skill is a working hypothesis, not a fixed rulebook. After each real paper application:
- Record where a recommendation failed, drew justified pushback, or conflicted with venue or paper constraints.
- If a rule produced a bad recommendation in context, do not silently follow it next time. Either:
- Narrow the rule (add the condition under which it applies), or
- Flip the rule's default (if the user's corrected choice is the better general default), or
- Remove the rule (if it was over-generalized guidance).
- Update
references/design-rationale.mdonly when the lesson is generalizable; do not preserve private conversations, project identifiers, or dated interaction logs. - User-originating preferences always outrank inferred rules from exemplars.
Signals to watch for:
- "Why did you put X in Y?" → the placement rule is probably over-normative.
- "I don't need to follow this, I'll do Z instead." → the default may be wrong for this user's venue/style.
- "This feels defensive / redundant / apologetic." → the rule is adding ceremony, not substance.
When in doubt, ask the user to ratify the update before applying it to the skill file.
Cross-skill handoff
After this skill produces a section draft or restructuring plan, hand off to paper-refinement-skills for sentence-level polish. Do not duplicate prose-polish rules here.
For substantial edits, also check that section-level claims still match the paper's tables, figures, citations, notation, and rendered cross-references.
What ships with it: 11 files
130.2 KB alongside SKILL.md
agents/
- openai.yaml275 B
checklists/
- reviewer-redflags.md26.2 KB
decision-trees/
- paper-type-router.md2.2 KB
references/
- design-rationale.md3.6 KB
sections/
- abstract.md13.0 KB
- conclusion.md10.6 KB
- experiments.md28.8 KB
- introduction.md18.7 KB
- method.md16.2 KB
- related-work.md9.7 KB
- LICENSE1.0 KB