All skills
imbad0202 avatar

/academic-paper

@825afa9

12-agent academic paper writing pipeline. 11 modes (full/plan/outline/revision/revision-coach/abstract/lit-review/format-convert/citation-check/disclosure/rebuttal-audit). 6 paper types, 5 citation formats, bilingual abstracts, LaTeX/DOCX-via-Pandoc/PDF output. Style Calibration + Writing Quality Check + Anti-Patterns with IRON RULE markers. Triggers: write paper, academic paper, guide my paper, parse reviews, I got reviewer comments, revision roadmap, should we push back, conference rebuttal, grant panel response, audit my rebuttal, check my response draft, AI disclosure, check citations, citation check, check my references, verify references, look over the refs, 寫論文, 學術論文, 引導我寫論文, 審查意見, 我收到審查意見, 修訂路線圖, 評估回覆, 檢查引用, 引用檢查, 檢查參考文獻, 核對文獻, 논문 작성, 초록 작성, 논문 수정, 논문 계획을 도와줘, 심사 의견을 받았어, 심사 의견 반영, 답변서 점검, AI 사용 고지, 인용 확인, 인용 형식 검사, enmendar mi artículo, redactar artículo, guía mi artículo, analizar reseñas, auditar mi respuesta, verificar borrador de respuesta, verificar citas, divulgación de IA.

Use this Skill: https://skilld.dev/gh/imbad0202/academic-research-skills/academic-paper

This session only. Nothing lands on disk.

examplesversion_family_reconciliation_example.md

≈982 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Version-Family Reconciliation Example (Kong #258)

This example shows how ARS surfaces a citation-version mismatch without auto-correcting the manuscript.

Scenario

The author wants to cite Attention Is All You Need. Their corpus contains two concrete versions:

  • arXiv preprint v1
  • NeurIPS 2017 proceedings record

The scholar selects the proceedings record as the primary citable version, but the draft sentence quotes the arXiv v1 locator while the reference list renders proceedings metadata.

Sidecar

phase2_investigation/version_records.yaml:

schema_version: "1.0"
version_records:
  - version_family_id: attention-transformer-family
    canonical_slug: vaswani2017
    primary_version_key: attention-neurips-2017
    discovery_status: user_confirmed
    reconciliation_note: "Proceedings record selected as default citation; arXiv v1 can be cited explicitly for preprint-specific claims."
    known_versions:
      - version_key: attention-arxiv-v1
        citation_key: vaswani2017-arxiv-v1
        kind: arxiv_preprint
        title: "Attention Is All You Need"
        year: 2017
        venue: "arXiv"
        doi: null
        arxiv_id: "1706.03762v1"
        url: "https://arxiv.org/abs/1706.03762v1"
        publication_date: {value: "2017-06", precision: month}
        metadata_provenance: arxiv_api
        source_locator: "arxiv:1706.03762v1"
        claim_scope_note: "Use only for text verified against the v1 preprint."
        notes: null
      - version_key: attention-neurips-2017
        citation_key: vaswani2017-neurips
        kind: proceedings
        title: "Attention Is All You Need"
        year: 2017
        venue: "NeurIPS 2017"
        doi: null
        arxiv_id: null
        url: "https://papers.nips.cc/paper/7181-attention-is-all-you-need"
        publication_date: {value: "2017", precision: year}
        metadata_provenance: user_override
        source_locator: "papers.nips.cc:7181"
        claim_scope_note: "Use for proceedings metadata and the final citable record."
        notes: null

Problem Draft

The preprint v1 states the Transformer architecture relies entirely on attention mechanisms (Vaswani et al., 2017).<!--ref:vaswani2017-neurips ok--><!--anchor:section:arxiv:1706.03762v1:sec3-->

The citation slug points to the proceedings record, but the prose and anchor point to arXiv v1.

Advisory

VERSION_INCONSISTENT_CITATION: citation metadata, locator, or quoted claim mixes multiple records in version_family_id=attention-transformer-family. The rendered slug vaswani2017-neurips belongs to attention-neurips-2017, while the prose/anchor refers to attention-arxiv-v1. Select one version or explicitly separate the claims.

Acceptable Fix A — Cite the Preprint Explicitly

The arXiv v1 preprint states the Transformer architecture relies entirely on attention mechanisms (Vaswani et al., 2017).<!--ref:vaswani2017-arxiv-v1 ok--><!--anchor:section:arxiv:1706.03762v1:sec3-->

Acceptable Fix B — Use the Proceedings Record

The proceedings version presents the Transformer architecture as relying entirely on attention mechanisms (Vaswani et al., 2017).<!--ref:vaswani2017-neurips ok--><!--anchor:section:papers.nips.cc:7181:sec3-->

Acceptable Fix C — Compare Versions Explicitly

The arXiv v1 preprint and the NeurIPS proceedings version both frame the Transformer as an attention-only architecture, but the manuscript cites the proceedings version as the primary record (Vaswani et al., 2017).<!--ref:vaswani2017-arxiv-v1 ok--><!--anchor:section:arxiv:1706.03762v1:sec3--><!--ref:vaswani2017-neurips ok--><!--anchor:section:papers.nips.cc:7181:sec3-->

The important rule is that each version-bound claim has its own slug and locator. ARS does not decide which version is "better"; the scholar selects the version appropriate to the claim.

Source: SKILL.md on GitHub

1 alert1d5 checks · Risk SAFE
  • Gen Agent Trust Hub1d

    This skill provides a highly structured 12-agent academic paper writing pipeline. It contains robust internal defenses, including explicit 'instruction-data' boundaries and sanitization of user-provided content. However, the skill's complexity, its reliance on shell command execution for document processing and utility scripts, and its heavy ingestion of untrusted third-party materials (such as reviewer comments and external research content) present a surface for potential misuse and indirect prompt injection.

  • Socket1d

    No alerts

  • Snyk1d

    Risk: LOW · No issues

  • Runlayer6mo

    7/44 files flagged

  • ZeroLeaks5mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at 825afa9. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 days ago.

Activeupdated last week
Other metadata
metadata
{
  "version": "3.3.1",
  "last_updated": "2026-08-15",
  "status": "active",
  "data_access_level": "raw",
  "task_type": "open-ended",
  "related_skills": [
    "deep-research",
    "academic-paper-reviewer",
    "academic-pipeline"
  ]
}

README badge

README badge for imbad0202/academic-research-skills/academic-paper