Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    BibTeX Cleaner

    1

    Clean a .bib file: duplicate keys, missing fields, malformed entries, and unused references.

    Free

    1 useSecurity scanned
    BibTeX Cleaner

    BibTeX Cleaner

    Example session with this skill installed

    Clean my bib file references.bib and check if any are unused in main.tex. Report all missing fields.

    • Read your context and instructions
    • Compiled the bibtex cleaner

    I have analyzed references.bib and cross-referenced it with main.tex. A cleaned file references.clean.bib has been generated, and the audit results are summarized below.

    BibTeX report — references.bib

    Entries: 24 | Duplicates: 1 | Missing fields: 3 | Unused: 2

    Audit Summary

    Issue TypeDetails
    Duplicate Keysdoe2021 appears on line 14 and line 102.
    Missing Fieldswang2023 (article): missing journal, year.
    Missing Fieldscite:miller2020 (inproceedings): missing booktitle.
    Unused Entriesarchived_tech_2015, survey_draft_v1.
    Malformed Bracesbrown2022: unclosed brace in title field.

    Cleaned Bibliography Preview

    @article{doe2021,
      author = {Doe, John},
      title = {Advances in Neural Systems},
      journal = {AI Journal},
      year = {2021},
      volume = {12},
      pages = {45-60}
    }
    
    @article{wang2023,
      author = {Wang, Li},
      title = {Data Processing Scales},
      % MISSING: journal, year
    }
    

    Next steps

    1. Review references.clean.bib and manually add the missing journal and year for wang2023.
    2. Delete the unused entry archived_tech_2015 if it is no longer needed for the final submission.
    3. Update your LaTeX preamble to point to the new references.clean.bib file.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Managing a bibliography for a large paper or thesis often leads to broken builds and inconsistent citations. This skill identifies malformed entries, resolves duplicate keys, and finds unused references that bloat your files. It ensures your BibTeX file is technically sound and ready for compilation without manual auditing.

    What it does

    • Duplicate resolution identifies identical keys across the file and preserves the entry with the most data.
    • Field validation checks every entry against required BibTeX standards to flag missing journals, years, or publishers.
    • Usage analysis cross-references your .bib with .tex or .typ files to isolate citations not actually used in the document.
    • Syntax repair detects and reports malformed braces or broken entry structures that prevent successful compilation.

    How it works

    1. Source analysis where the skill reads your .bib file and any associated LaTeX or Typst source files.
    2. Validation run using a cleaning script to generate a cleaned .bib copy and a detailed Markdown report.
    3. Conflict resolution where duplicates are merged or renamed based on data richness.
    4. Final verification through a report review to address missing fields manually before the next build.

    Frameworks & tools

    This skill uses a Python-based cleaning utility and a standard BibTeX checklist to validate entry types such as @article, @book, and @inproceedings. It is compatible with LaTeX and Typst workflows.

    Why this beats prompting it yourself

    Generic LLM prompts often hallucinate missing DOI numbers or change citation keys, which breaks your in-text citations. This skill follows strict guardrails to never fabricate data and uses a systematic script-based approach to keep the file technically consistent.

    • Parsing is heuristic; unusual BibTeX syntax may need manual fixes.
    • Does not verify that metadata is true, only that fields exist.
    • Does not reformat citation style; that belongs to the bibliography style.
    • @string and @preamble blocks are ignored, not validated.
    • Unused detection depends on the source files you provide.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 27 days ago

    • Free with an Agensi account

    Requires

    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files

    File Scopes

    .bib
    .tex
    .typ
    Listed27 days ago

    What's inside

    Frequently Asked Questions

    The skill identifies duplicate entry keys, checks for missing required fields (like DOIs or publication years), and compares the bibliography against your manuscript to flag unused references that clutter your file.

    It is compatible with standard LaTeX (.tex) files and the modern Typst (.typ) format to cross-reference citations and ensure your bibliography matches your actual manuscript usage.

    No, this skill focuses on data integrity and formatting; it will never hallucinate or invent missing data like DOIs or abstracts to ensure your academic citations remain factually accurate.

    You receive the core Python-based cleaning script, the schema validation checklist, and the structured workflow for analyzing, reporting, and sanitizing your bibliography files.

    The skill generates a new, sanitized version of your bibliography and provides a detailed report of all changes, ensuring your original source file remains untouched and safe.

    New in Technical Documentation