Bibverify

Verify bibliographic existence and metadata consistency without destructive edits.

Chinese · Install · Quick start · MCP · Development

Bibverify is a BibTeX metadata verification tool for researchers, editors, automation, and AI assistants. It starts with exact DOI, PMID, PMCID, or arXiv identifiers and then scores candidates from title, author, year, venue, and pagination evidence.

Bibverify evaluates whether a bibliographic record can be located in the queried sources and whether its metadata agrees. It does not prove that research findings are true, data is authentic, or a venue is reputable. A missing database record is not evidence that a reference is fabricated. By default, the source .bib file is never overwritten.

Highlights

• Identifier-first lookup for DOI, PMID, PMCID, and arXiv IDs.
• Crossref, OpenAlex, Semantic Scholar, PubMed, Europe PMC, CORE, DBLP, arXiv, bioRxiv, and more.
• Explainable multi-signal matching across identifiers, title, authors, year, venue, and pages. A resolvable DOI with a materially different title becomes identifier_conflict instead of being hidden by title search.
• Structured provider outcomes distinguish a genuine no-match from rate limiting, authentication, network, parsing, and provider failures.
• Non-destructive merging preserves abstract, keywords, file, note, and custom fields that providers do not return; conflicting persistent identifiers are never overwritten automatically.
• A shared connection pool with retries, exponential backoff, and Retry-After support for 429/5xx responses.
• An expiring SQLite cache for successful GET responses; failures are never cached.
• Cross-platform path, encoding, Unicode filename, UTF-8 BOM, and CRLF handling.
• JSON output, stable exit codes, a Python API, and an MCP server built on the official SDK.
• Atomic output writes and byte-preserving backups.

Requirements

• Windows, macOS, or Linux
• Network access to the enabled metadata APIs
• Python 3.11–3.14 when using the Python distribution; npm, containers, and native packages bundle their runtime

GitHub Actions tests all three operating systems across all four supported Python versions.

Install

The current published release is v0.3.0. This branch prepares v0.4.0; commands marked v0.4.0 become usable only after their linked npm, GHCR, or GitHub Release artifact has been published.

Run without a permanent installation:

uvx bibverify --version

Starting with v0.4.0, Node.js users can use the zero-dependency npm launcher. It downloads the matching native release, verifies SHA256SUMS, and forwards every argument and exit code:

npx --yes @hylouis233/bibverify --version
pnpm dlx @hylouis233/bibverify --version
bunx --bun @hylouis233/bibverify --version

For a persistent CLI in an isolated Python environment, use uv tool or pipx:

uv tool install bibverify

pipx install bibverify

Inside a virtual environment, regular pip also works:

python -m pip install --upgrade bibverify

Each release also provides smoke-tested native packages on GitHub Releases. Beginning with v0.4.0, the release matrix covers Windows x64 plus macOS and glibc 2.28+ Linux on both x64 and ARM64. musl-based Linux users should use the Python package or container image. Windows ARM64 is not published natively yet because an MCP runtime dependency does not currently provide Windows ARM64 wheels; npm automatically uses the tested x64 build under Windows 11 emulation, or you can use the ARM64 container instead.

The same release publishes a multi-architecture container:

docker run --rm ghcr.io/hylouis233/bibverify:0.4.0 --version

Package-manager manifests are generated from the final release bytes, not from unverified build inputs. Homebrew, Scoop, and WinGet are not live catalog entries yet: v0.4.0 will attach submission- ready manifests to GitHub Release, after which each external catalog still requires onboarding or review. See Distribution channels for rollout status.

Quick start

Convert a DOI to BibTeX

bibverify doi 10.1038/nature12373 --key example2013

For a machine-readable response:

bibverify doi 10.1038/nature12373 --json

Verify a .bib file

Create a starter configuration:

bibverify config init

Place references.bib next to the configuration and run:

bibverify check --config config.json

You can override the input and output paths from the command line:

bibverify check references.bib --config config.json --output-dir bibverify-output

Inspect results without writing any file:

bibverify check references.bib --dry-run --json

After reviewing the report, explicitly apply high-confidence field updates. Bibverify creates a byte-for-byte backup first:

bibverify check references.bib --apply

PowerShell example:

py -m bibverify check '.\Bibliography\references.bib' --output-dir '.\Verification results'

The v0.2 forms remain available for compatibility, although new scripts should use subcommands:

bibverify config.json
bibverify --doi 10.1038/nature12373 --key example2013

Configuration

A minimal configuration looks like this:

{
  "language": "EN",
  "bib_file": "references.bib",
  "encoding": "auto",
  "output_dir": "bibverify-output",
  "user_info": {
    "email": "your_email@example.com",
    "app_name": "Bibverify"
  }
}

See config_template.json for every commonly used option.

Path behavior is intentionally predictable:

• Relative bib_file and output_dir values are resolved from the directory containing config.json, not the shell's current directory.
• If output_dir is omitted, output is written next to the input bibliography.
• encoding: "auto" tries UTF-8 with BOM, UTF-8, and GB18030. It does not fall back to Latin-1 and silently turn unknown bytes into mojibake.

API keys and email

Keys can be stored in a local configuration, but environment variables are safer and harder to commit accidentally:

PowerShell:

$env:BIBVERIFY_EMAIL = 'you@example.com'
$env:BIBVERIFY_OPENALEX_API_KEY = '...'
bibverify check --config config.json

Bash or Zsh:

export BIBVERIFY_EMAIL='you@example.com'
export BIBVERIFY_OPENALEX_API_KEY='...'
bibverify check --config config.json

Query and matching settings

{
  "query_settings": {
    "delay_between_requests": 0.5,
    "timeout": 10,
    "connect_timeout": 3.05,
    "read_timeout": 20,
    "max_retries": 3,
    "backoff_factor": 0.5,
    "stop_on_first_match": true,
    "match_threshold": 0.86,
    "ambiguous_threshold": 0.68,
    "auto_update_threshold": 0.92,
    "cache_enabled": true,
    "cache_ttl_hours": 168,
    "cache_path": ".bibverify-cache.sqlite3"
  }
}

connect_timeout and read_timeout separately bound connection setup and response reads; the compatibility timeout field remains available. match_threshold controls automatic candidate acceptance, ambiguous_threshold controls which plausible candidates enter review, and auto_update_threshold is an additional gate for field changes. Higher values are more conservative. Relative cache paths are resolved from the config directory.

The official bioRxiv details route does not provide arbitrary title search. Bibverify therefore calls bioRxiv only for an exact 10.1101/... DOI and leaves title-only discovery to providers whose contracts support it, such as Crossref and Europe PMC.

CLI reference

bibverify check [BIB_FILE] [--config PATH] [--output-dir DIR] [--format txt|json|jsonl|csv] [--dry-run|--apply] [--json]
bibverify doi DOI [--key KEY] [--config PATH] [--json]
bibverify config init [--output PATH] [--force]
bibverify doctor [--config PATH] [--json]
bibverify providers list [--json]
bibverify cache clear [--config PATH]
bibverify benchmark [--dataset PATH]
bibverify mcp [--config PATH] [--workspace-root DIR] [--transport stdio|streamable-http]
bibverify agent init [--target generic|codex|claude|cursor] [--output PATH]
bibverify skill export [--target ...]

Exit codes:

With --json, stdout contains JSON only. Diagnostics go to stderr, which keeps the command safe for CI and scripts.

Output files