check-refs
Skill39 starsChanged 8 months agoMITView on GitHub
Match references in a LaTeX paper against academic databases, flag entries that need review, and suggest missing DOIs or URLs. Use when checking citations or preparing a paper for submission.
In this repository12 other files it gives its agents.
- accessible-pdfSkill
- anonymize-repoSkill
- arxiv-metadataSkill
- arxiv-prepSkill
- bib-doi-toggleSkill
- critique-figuresSkill
- critique-manuscriptSkill
- doi-bibtexSkill
- latex-cleanupSkill
- openalexSkill
- presubmit-checksSkill
- verify-mathSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
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
- Distinctive: line 13
Use user-provided paths when available.
- Distinctive: line 15
Run the database-only check by default:
- Very specific: line 23
Read the generated bibsleuth-report.md.
- Very specific: line 31
If the user explicitly requested full analysis at the outset, do not force a redundant --no-llm pass first.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.
Files
Check citation references
Match a LaTeX paper's references against academic databases with bibsleuth. Database matches support bibliographic verification but do not prove that a citation supports the manuscript's claim.
Workflow
Use user-provided paths when available. Otherwise, identify the manuscript entrypoint from the project's build configuration and TeX structure, then resolve its bibliography declarations. Ask only if multiple plausible papers remain.
Run the database-only check by default:
uvx bibsleuth check <TEX_FILE> --bib <BIB_FILE> --no-llmuvx supplies the tool without a separate installation step. Allow enough time for large bibliographies and use asynchronous execution when the harness supports it.
Read the generated bibsleuth-report.md. Prioritize unverified and error results, summarize suggested DOI or URL patches, and link the full report. Apply patches only when the user requested reference repair or approves the proposed edits, then check the diff for unintended bibliography changes.
For claim-level analysis, missing-citation suggestions, and contradiction searches, run the LLM mode only when the user requests it and accepts the additional model-provider exposure:
uvx bibsleuth check <TEX_FILE> --bib <BIB_FILE>If the user explicitly requested full analysis at the outset, do not force a redundant --no-llm pass first. Present suspected miscitation and literature findings as items for review, not established errors.
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
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.