agentleFS
joshua-zyyjoshua-zyy/academic-paper-writer/skills/academic-citation/SKILL.md

academic-citation

Skill111 starsChanged 5 months agoMITView on GitHub

Search, verify, and map citations for CS/AI/ML papers. Produces VERIFIED/UNVERIFIED reference lists with Citation-to-Claim maps and Exemplar Sets. Use when: finding references for a paper section, verifying citation accuracy, building exemplar sets for introduction/related work learning, checking if existing citations are real and accurate, supplementing local literature library. Triggers on: 找引用, 文献检索, citation pass, find references, reference check, 补文献, citation verification, Exemplar Set, search papers, verify citation, 核验文献, 查引用, literature search, reference verification, citation verification with reading, 引用确认, 全文阅读验证.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read manifest.yaml. It declares always_load files, axes, and references.on_demand.

    line 12
  2. Very specific:

    Read every file listed under always_load.

    line 13
  3. Distinctive:

    If the user wants to use a local literature library, first distinguish:

    line 19

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Academic Citation

将此 skill 视为"文献取证代理",而不是搜索结果搬运器。

Router Protocol

  1. Read manifest.yaml. It declares always_load files, axes, and references.on_demand.
  2. Read every file listed under always_load. These are the skill's binding rules — not reference material.
  3. Apply the loaded material as constraints:
    • stance.md defines non-negotiable rules, source/inference boundary, and scope.
    • red-lines.md defines absolute prohibitions. Do not negotiate these.
    • output-contract.md defines deliverables per mode.
    • anti-patterns.md defines known failure modes and their correct alternatives.
  4. If the user wants to use a local literature library, first distinguish:
    • local_ref_md_dir: an existing Markdown library containing _index_ref.json.
    • local_ref_pdf_dir: a PDF directory, or any local library directory that lacks _index_ref.json.
  5. For local_ref_pdf_dir, stop before citation search. Tell the user to register a MinerU API token, set MINERU_API_TOKEN, run scripts/convert-pdfs-to-md.py, and return only after they confirm PDF→MD completion. Do not proceed to local reading until the generated MD directory has _index_ref.json.
  6. Detect the workflow using the manifest's workflow axis. Literature reading output must separate source quotes from agent inference; only source: 原文 or source: 图片 content may be used as citation evidence.
  7. Echo the selected mode to the user before executing.
  8. Reach for references/ only when the manifest's references.on_demand condition is satisfied.

Modes

ModeUse when
full-citation-passComplete coverage for a full paper or core section
targeted-citation-searchSpecific claim, section, or topic
exemplar-set-onlyBuild Exemplar Set only, not full citation list
citation-verificationVerify metadata of existing candidate list
local-citation-passLocal MD library first, then web supplement
citation-verification-with-readingFull-text reading after paper draft completion

Agent Dispatch

AgentPurpose
agents/citation_agent.mdLiterature search strategy (4-class query templates, output schema)
agents/literature-reader-agent.mdLiterature reading & extraction (MD full-text → LiteratureReadingReport)

Dispatched by academic-paper-writer orchestrator at Step 3. Subagents search and verify only; they must not modify project files or write paper prose independently. literature-reader-agent is dispatched in parallel by citation_agent or the orchestrator at Step 3a/3b.

Independent Use

InputModePriorityBehavior
Section + keywordsfull-citation-pass1 (explicit)Full 6-step workflow
Citation list / seed paperscitation-verification1 (explicit)Metadata only, no extra search
local_ref_md_dir provided and _index_ref.json existslocal-citation-pass2 (path trigger)Local MD + batch parallel reader agent dispatch
Local PDF library path provided, or local path lacks _index_ref.jsonpdf2md-preflight1 (blocking preflight)Instruct MinerU setup and script command; wait for user confirmation
Section onlytargeted-citation-search3 (single feature)Auto-keyword, 4-class query
Explicit Exemplar Set requestexemplar-set-only1 (explicit)Exemplar Set only

Multi-condition: lower priority number wins.

ScenarioRecommended
Just searching/verifying citationsThis skill (standalone)
Integrating into paper draftingacademic-paper-writer orchestrator
Existing draft needs citation reviewThis skill → academic-reviser

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.