Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add AutoConference/AutoConference-skill --skill paper-writinggit clone --depth 1 https://github.com/AutoConference/AutoConference-skillWrote 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.
[](https://agentmods.dev/skills/autoconference/autoconference-skill/paper-writing)<a href="https://agentmods.dev/skills/autoconference/autoconference-skill/paper-writing"><img src="https://agentmods.dev/badge/skills/autoconference/autoconference-skill/paper-writing/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/autoconference/autoconference-skill/paper-writing"><img src="https://agentmods.dev/badge/skills/autoconference/autoconference-skill/paper-writing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00070 | $0.03418 |
| Opus 5.5 | $0.00028 | $0.01367 |
| Sonnet 5.5 | $0.00014 | $0.00684 |
| Haiku 4.5 | $0.00007 | $0.00342 |
Grade A, and why
paper-writing 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 284 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Writing a paper that survives being checked
You turn evidence into a manuscript. You do not produce evidence, and you do not
decide what it supports. Those belong to whatever runs the experiments; the line
between you is ../interfaces/evidence-interface.md, and it is not yours to move.
A paper is not finished when it reads well. It is finished when it reads well and five gates pass on the compiled artifact. Structure, tables, equations, citations, prose patterns, and the rendered page itself are all mechanically checkable, so they are checked rather than remembered.
The contract
references/paper-writing-template.md is the layout contract. It is measured
from the complete arXiv LaTeX source of arXiv:2506.21656v3, not estimated from a
rendered page, and it is not optional. Load its companions as the work reaches
them:
| File | Load when |
|---|---|
references/paper-writing-template.md |
always, first |
../interfaces/evidence-interface.md |
reading aggregates or a readiness verdict |
references/section-playbook.md |
drafting or revising any section |
references/table-grammar.md |
creating or editing any table |
references/equation-grammar.md |
writing the method or any display math |
references/figure-ladder.md |
deciding what visuals the evidence supports |
references/figure-icons.md |
filling the system figure, or changing one of its glyphs |
references/prose-rules.md |
any manuscript-facing prose pass |
references/audit-protocol.md |
the independent review round |
references/source-selection.md |
maintaining this skill |
Setup
The template under template/ is a family, not one scaffold. Sixteen
design axes — layout, opening figure, contribution list, related-work
placement, method opening, equation register, experiments organization,
table grammar, ablation grammar, secondary figure, appendix, limitations,
devices, palette, system figure, title mark — each take values measured from
one of seven real papers (the palette values are published colour schemes)
(references/design-axes.md). scripts/design_paper.py draws one coherent
combination for a seed, records it in paper.json under design, assembles
paper/ from the drawn blocks, and writes DESIGN-BRIEF.md stating the
shape every section must take. Every gate then checks the manuscript against
that draw, not against a single exemplar.
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- config.example.json 238 B
- references/audit-protocol.md 7.7 KB
- references/design-axes.md 24 KB
- references/equation-grammar.md 5.5 KB
- references/figure-icons.md 11 KB
- references/figure-icons.verified.txt 9.8 KB
- references/figure-ladder.md 7.1 KB
- references/paper-writing-template.md 17 KB
- references/prose-rules.md 7.0 KB
- references/section-playbook.md 8.5 KB
- references/source-selection.md 7.5 KB
- references/table-grammar.md 21 KB
- scripts/build_paper.sh 2.6 KB runs code
- scripts/check_citations.py 18 KB runs code
- scripts/check_paper_structure.py 70 KB runs code
- scripts/check_prose_quality.py 15 KB runs code
- scripts/check_render.py 11 KB runs code
- scripts/design_paper.py 45 KB runs code
- scripts/gate.sh 3.5 KB runs code
- scripts/make_paper_data.py 37 KB runs code
- scripts/render_paper.sh 776 B runs code
- template/.gitignore 39 B
- template/blocks/ablation/cumulative.tex 1.3 KB
- template/blocks/ablation/default-delta.tex 1.6 KB
- template/blocks/ablation/naive-vs-ours.tex 1003 B
- template/blocks/ablation/none.tex 130 B
- template/blocks/ablation/paired-minipage.tex 1.6 KB
- template/blocks/ablation/wo-rows.tex 1.3 KB
- template/blocks/appendix/deep.tex 1.4 KB
- template/blocks/appendix/lettered.tex 962 B
- template/blocks/appendix/none.tex 193 B
- template/blocks/appendix/single.tex 607 B
- template/blocks/appendix/supp-toc.tex 1.6 KB
- template/blocks/contributions/bullets.tex 400 B
- template/blocks/contributions/inline.tex 348 B
- template/blocks/contributions/none.tex 375 B
- template/blocks/contributions/numbered.tex 389 B
- template/blocks/devices/algorithm-float.tex 885 B
- template/blocks/devices/data-card.tex 780 B
- template/blocks/devices/efficiency-scatter.tex 1.4 KB
- template/blocks/devices/llm-usage.tex 395 B
- template/blocks/devices/principle.tex 397 B
- template/blocks/devices/prior-limitations.tex 337 B
- template/blocks/devices/research-questions.tex 546 B
- template/blocks/devices/rhetorical-hook.tex 260 B
- template/blocks/devices/symbols-table.tex 455 B
- template/blocks/experiments/04b_ablation_section.tex 902 B
- template/blocks/experiments/bold-lead-stream.tex 1.3 KB
- template/blocks/experiments/by-task.tex 1.6 KB
- template/blocks/experiments/qual-quant-ablation.tex 1.9 KB
- template/blocks/experiments/roadmap-setup-results.tex 1.5 KB
- template/blocks/experiments/separate-ablation-section.tex 968 B
- template/blocks/experiments/setup-ablation-sota.tex 1.4 KB
- template/blocks/experiments/setup-main-qual-ablation.tex 1.6 KB
- template/blocks/limitations/absent.tex 99 B
- template/blocks/limitations/appendix.tex 357 B
- template/blocks/limitations/conclusion-paragraph.tex 310 B
- template/blocks/limitations/own-section.tex 298 B
- template/blocks/method-open/components.tex 874 B
- template/blocks/method-open/formal.tex 670 B
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.
- 1mo ago Changed · +9 lines 97ac887e4528
- 1mo ago First seen · 275 lines · 70 tokens per session scan A b6257deebfce
paper-writing is a skill published in the GitHub repository AutoConference/AutoConference-skill (13 stars, last pushed 4d ago), licensed Apache-2.0. It adds 70 tokens to every session and 3,418 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-09-24.
paper-writing compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| paper-writingAutoConference/AutoConference-skill | Skill | 13 | 4d ago | Apache-2.0 | A | 70 | Claude Code, Codex |
| instrument-data-to-allotropeanthropics/knowledge-work-plugins | Skill | 36k | today | Apache-2.0 | A | 123 | Claude Code |
| exploratory-data-analysisK-Dense-AI/scientific-agent-skills | Skill | 57k | 4d ago | MIT | A | 83 | Claude Code |
| matlabK-Dense-AI/scientific-agent-skills | Skill | 57k | 4d ago | MIT | A | 44 | Claude Code |
| phylogeneticsK-Dense-AI/scientific-agent-skills | Skill | 57k | 4d ago | MIT | A | 71 | Claude Code, Codex |
| research-engineerdavila7/claude-code-templates | Skill | 33k | yesterday | MIT | A | 43 | Claude Code, Codex |
| mapping-to-snomedmaziyarpanahi/openmed | Skill | 5.8k | 2d ago | Apache-2.0 | A | 205 | Claude Code |