Is the Check Citations skill free?

Yes. The skill itself is free from PHY041/claude-skill-citation-checker. SkillProof publishes the install command and an independent test verdict at no cost.

What is the Check Citations SkillProof Score?

8.8/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 4/5.

How do I install Check Citations?

Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.

Can I use Check Citations with Cursor, Copilot, Gemini CLI, Codex or other AI tools?

The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.

Check Citations

Live 3-database check that caught a fabricated and a chimeric citation

by PHY041 · PHY041/claude-skill-citation-checker

Tested · Works

Install — copy the command Download source ↗

Test report

Verdict: Tested · Works
Score: 8.8/10
Tested: Jul 17, 2026
Environment: Claude Code 2.x (agent harness)
Upstream re-checked: Aug 10, 2026 · 21ba14c

Ran the bundled checker live on a 5-entry .bib seeded with one fabricated paper and one chimeric citation (real GPT-3 title, fake authors): the fabricated entry came back 0/3 databases with file and line number, and the chimeric one was flagged 'title matches but authors don't'. The core guarantee held — no fake was ever marked verified — while a memory-only baseline could not prove the plausible fake wrong at all. It did flag the real BERT paper as not found (the exact DOI-less false positive its own docs admit), and --json mode is broken as shipped: progress lines contaminate stdout, so the documented pre-commit hook crashes on json.load. Expect roughly 80 seconds per 5 citations due to rate-limit sleeps.

Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works

Installs cleanly 5/5
Triggers reliably 5/5
Output vs. baseline 8/10
Docs & honesty 4/5

What Check Citations does

Verifies every entry in a .bib file against CrossRef, Semantic Scholar, and OpenAlex to catch AI-hallucinated references before submission. Flags fully fabricated papers (found in 0 databases), chimeric citations (real title, wrong authors, detected via zero author overlap), and red-flag metadata like invalid DOIs or future years. Triggers when reviewing bibliographies, AI-written literature sections, or LaTeX manuscripts pre-submission.

How to install Check Citations

git clone https://github.com/PHY041/claude-skill-citation-checker
mkdir -p ~/.claude/skills
cp -r claude-skill-citation-checker ~/.claude/skills/check-citations
pip3 install requests

Skills live in ~/.claude/skills/ (global) or .claude/skills/ (per-project). Restart Claude Code after installing.

Commands — how to trigger Check Citations

/check-citations Live 3-database check that caught a fabricated and a chimeric citation

It also activates on plain-language prompts like these:

Check my references.bib for fake or hallucinated citations
Verify every DOI in this bibliography against CrossRef and OpenAlex
Did my AI-written lit review section cite any papers that don't exist?

Frequently asked questions

Is the Check Citations skill free?: Yes. The skill itself is free from PHY041/claude-skill-citation-checker. SkillProof publishes the install command and an independent test verdict at no cost.
Does Check Citations work with Claude Code?: We tested it with Claude Code 2.x (agent harness) on Jul 17, 2026. Verdict: Tested · Works. Ran the bundled checker live on a 5-entry .bib seeded with one fabricated paper and one chimeric citation (real GPT-3 title, fake authors): the fabricated entry came back 0/3 databases with file and line number, and the chimeric one was flagged 'title matches but authors don't'. The core guarantee held — no fake was ever marked verified — while a memory-only baseline could not prove the plausible fake wrong at all. It did flag the real BERT paper as not found (the exact DOI-less false positive its own docs admit), and --json mode is broken as shipped: progress lines contaminate stdout, so the documented pre-commit hook crashes on json.load. Expect roughly 80 seconds per 5 citations due to rate-limit sleeps.
What is the Check Citations SkillProof Score?: 8.8/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 4/5.
How do I install Check Citations?: Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
Can I use Check Citations with Cursor, Copilot, Gemini CLI, Codex or other AI tools?: The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.