Skip to content
SKILLS
Agent SkillPublisher identity verifiedNo findings from static checks

@spboucher/citation-verificationv1.0.0

Verify citations against Crossref, OpenAlex, arXiv and doi.org: a DOI must resolve and its metadata must match. Audits .bib files, resolves references without DOI, flags retractions, guides literature searches.

  • downloadable
  • installable
  • executable
Version
1.0.0
Published
Nov 10, 2026
Licence
MIT
Archive
40 kB
Files
16

Install

Install with the Skills CLI (verifies the checksum, previews the destination, never runs scripts)

skills install @spboucher/citation-verification@1.0.0 --client claude-code

Add --project to install into .claude/skills/citation-verification instead of ~/.claude/skills/citation-verification.

Installs to (user)
~/.claude/skills/citation-verification
With --project
.claude/skills/citation-verification
Without the CLI: download, verify, extract (4 steps)
  1. 1Or manually: download the release archive

    curl -fsSLo citation-verification-1.0.0.tar.gz https://skills.spboucher.ai/api/v1/artifacts/spboucher/citation-verification/versions/1.0.0/download
  2. 2Verify the SHA-256 checksum

    echo "0b88da37115961e37cb68baa35c3cba1d6a081d2ba916490092403be782b9065  citation-verification-1.0.0.tar.gz" | shasum -a 256 -c -
  3. 3Inspect the contents before extracting

    tar -tzf citation-verification-1.0.0.tar.gz
  4. 4Extract into ~/.claude/skills/citation-verification

    mkdir -p ~/.claude/skills/citation-verification && tar -xzf citation-verification-1.0.0.tar.gz -C ~/.claude/skills/citation-verification
  • Declared permissions: network, filesystem:read, filesystem:write, shell, env.
  • Skills can contain scripts. Installing only copies files; read SKILL.md and any scripts before letting an agent use them.

Installing only copies or configures. Nothing is executed for you. The Skills CLI verifies the checksum before writing files.

Documentation

citation-verification

An agent skill for checking references instead of trusting them. It teaches the rule that matters (a DOI must resolve and its registered metadata must match the citation) and ships a command-line tool that applies it to one citation or a whole bibliography.

shell
node scripts/scholarly.mjs verify-bib references.bib --format md --out audit.md
node scripts/scholarly.mjs verify-doi 10.1016/0304-4076(86)90063-1 --title "Generalized autoregressive conditional heteroskedasticity" --year 1986
node scripts/scholarly.mjs retractions --bib references.bib
node scripts/scholarly.mjs help

What is inside

  • SKILL.md — the rules, the statuses and the workflows (audit a manuscript, add a reference, preprints, retractions, literature search).
  • scripts/scholarly.mjs — the tool. Node.js 20+, no dependency, JSON output.
  • scripts/lib/ — the engine, identical to the one in @spboucher/scholarly-mcp.
  • references/ — thresholds and status definitions, source coverage and limits, report template.

Requirements

Node.js 20 or newer and network access to Crossref, OpenAlex, arXiv and doi.org. No account. Optional environment variables: OPENALEX_API_KEY, SCHOLARLY_MAILTO, SCHOLARLY_ROOT, SCHOLARLY_TIMEOUT_MS.

Privacy

Requests carry a generic client name. The titles, DOIs and author names you check are sent to the four services; nothing else is. No contact address is sent unless you set SCHOLARLY_MAILTO yourself.

Limits

  • Comparisons are string heuristics with documented thresholds; every verdict comes with the claimed and registered values so you can overrule it.
  • Books, theses, reports and many regional journals are poorly covered: expect not_found, which does not mean invented.
  • It cannot tell whether a cited work supports the claim citing it.
  • Without an OpenAlex key, citation-graph commands share a small daily budget per IP address. Verification does not use OpenAlex.
  • Not tested: references in non-Latin scripts, bibliographies above a few dozen entries in one run, BibLaTeX-only constructs such as crossref inheritance and @xdata.

MIT licence.

Files

16 files · 120 kB unpacked

Select a file to preview it. Files are shown as plain text and never executed.

Security & permissions

Declared permissions

  • network
  • filesystem:read
  • filesystem:write
  • shell
  • env

Static checks

No advisory findings from static-1.0.0 on Nov 10, 2026. 10 informational notes.

Static checks are advisory. They never certify that a package is safe. Compatibility entries are publisher claims unless marked verified.

Compatibility

  • claude-code · claimed
  • codex · claimed
  • cursor · claimed
  • generic-agent-skills · claimed

Versions

1 release · immutable
VersionPublishedSizeSHA-256State
1.0.0latestNov 10, 202640 kB0b88da371159…published

Publication history

  1. release.publish@spboucher/citation-verification@1.0.0by Simon-Pierre Boucher
  2. artifact.create@spboucher/citation-verificationby Simon-Pierre Boucher
  • Agent SkillVerified publisher

    @spboucher/revision-detector-experiment

    Run a sequential revision experiment on a text and record what the Pangram detector reports for each version, as an observed outcome: frozen editorial plan, one submission per step, real evaluations, preservation checks, chart and report.

    Latest version
    v1.0.0
    Licence
    MIT
    Updated
    1 h ago
  • Agent SkillVerified publisher

    @spboucher/manuscript-peer-review

    Peer-review protocol for LaTeX and Markdown manuscripts: deterministic audit script, five review roles, adversarial verification of findings, anchored rubric, referee report and decision letter templates.

    Latest version
    v1.0.0
    Licence
    MIT
    Updated
    2 h ago
  • Agent SkillVerified publisher

    @spboucher/excel-model-workbook

    Build, audit and grade Excel model workbooks: layout standard, dependency-free command-line auditor (errors, hard-coded numbers, broken patterns, circular references, controls, stale values) and rubric-based grading aid.

    Latest version
    v1.0.0
    Licence
    MIT
    Updated
    2 h ago

↑↓ to navigate · enter to open · esc to close