{"id":"6d80d634-a19c-4271-8029-416e8e8c47c7","shortId":"J8ZJ6q","kind":"skill","title":"bib-verify","tagline":"Verify a BibTeX file for hallucinated or fabricated references by cross-checking every entry against CrossRef, arXiv, and DBLP. Reports each reference as verified, suspect, or not found, with field-level mismatch details (title, authors, year, DOI). Use when the user wants to che","description":"# BibTeX Verification Skill\n\nCheck every entry in a `.bib` file against real academic databases using the\nOpenJudge `PaperReviewPipeline` in BibTeX-only mode:\n\n1. **Parse** — extract all entries from the `.bib` file\n2. **Lookup** — query CrossRef, arXiv, and DBLP for each reference\n3. **Match** — compare title, authors, year, and DOI\n4. **Report** — flag each entry as `verified`, `suspect`, or `not_found`\n\n## Prerequisites\n\n```bash\npip install py-openjudge litellm\n```\n\n## Gather from user before running\n\n| Info | Required? | Notes |\n|------|-----------|-------|\n| BibTeX file path | Yes | `.bib` file to verify |\n| CrossRef email | No | Improves CrossRef API rate limits |\n\n## Quick start\n\n```bash\n# Verify a standalone .bib file\npython -m cookbooks.paper_review --bib_only references.bib\n\n# With CrossRef email for better rate limits\npython -m cookbooks.paper_review --bib_only references.bib --email your@email.com\n\n# Save report to a custom path\npython -m cookbooks.paper_review --bib_only references.bib \\\n  --email your@email.com --output bib_report.md\n```\n\n## Relevant options\n\n| Flag | Default | Description |\n|------|---------|-------------|\n| `--bib_only` | — | Path to `.bib` file (required for standalone verification) |\n| `--email` | — | CrossRef mailto — improves rate limits, recommended |\n| `--output` | auto | Output `.md` report path |\n| `--language` | `en` | Report language: `en` or `zh` |\n\n## Interpreting results\n\nEach reference entry is assigned one of three statuses:\n\n| Status | Meaning |\n|--------|---------|\n| `verified` | Found in CrossRef / arXiv / DBLP with matching fields |\n| `suspect` | Title or authors do not match any real paper — likely hallucinated or mis-cited |\n| `not_found` | No match in any database — treat as fabricated |\n\n**Field-level details** are shown for `suspect` entries:\n- `title_match` — whether the title matches a real paper\n- `author_match` — whether the author list matches\n- `year_match` — whether the publication year is correct\n- `doi_match` — whether the DOI resolves to the right paper\n\n## Additional resources\n\n- Full pipeline options: [../paper-review/reference.md](../paper-review/reference.md)\n- Combined PDF review + BibTeX verification: [../paper-review/SKILL.md](../paper-review/SKILL.md)","tags":["bib","verify","openjudge","agentscope-ai","agent","agent-skills","ai-agent","alignment","evaluation","grader","llm","reward"],"capabilities":["skill","source-agentscope-ai","skill-bib-verify","topic-agent","topic-agent-skills","topic-ai-agent","topic-alignment","topic-evaluation","topic-grader","topic-llm","topic-reward","topic-reward-model","topic-rlhf","topic-skill-md","topic-skills"],"categories":["OpenJudge"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/agentscope-ai/OpenJudge/bib-verify","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add agentscope-ai/OpenJudge","source_repo":"https://github.com/agentscope-ai/OpenJudge","install_from":"skills.sh"}},"qualityScore":"0.700","qualityRationale":"deterministic score 0.70 from registry signals: · indexed on github topic:agent-skills · 585 github stars · SKILL.md body (2,264 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-02T18:53:08.010Z","embedding":null,"createdAt":"2026-04-18T21:57:25.913Z","updatedAt":"2026-05-02T18:53:08.010Z","lastSeenAt":"2026-05-02T18:53:08.010Z","tsv":"'/paper-review/reference.md':322,323 '/paper-review/skill.md':329,330 '1':73 '2':82 '3':92 '4':100 'academ':62 'addit':317 'api':140 'arxiv':21,86,243 'assign':232 'author':40,96,251,292,296 'auto':214 'bash':112,145 'better':162 'bib':2,58,80,131,149,155,169,184,196,200 'bib-verifi':1 'bib_report.md':190 'bibtex':6,50,70,127,327 'bibtex-on':69 'che':49 'check':16,53 'cite':263 'combin':324 'compar':94 'cookbooks.paper':153,167,182 'correct':306 'cross':15 'cross-check':14 'crossref':20,85,135,139,159,207,242 'custom':178 'databas':63,270 'dblp':23,88,244 'default':194 'descript':195 'detail':38,277 'doi':42,99,307,311 'email':136,160,172,187,206 'en':220,223 'entri':18,55,77,104,230,282 'everi':17,54 'extract':75 'fabric':11,273 'field':35,247,275 'field-level':34,274 'file':7,59,81,128,132,150,201 'flag':102,193 'found':32,110,240,265 'full':319 'gather':119 'hallucin':9,259 'improv':138,209 'info':124 'instal':114 'interpret':226 'languag':219,222 'level':36,276 'like':258 'limit':142,164,211 'list':297 'litellm':118 'lookup':83 'm':152,166,181 'mailto':208 'match':93,246,254,267,284,288,293,298,300,308 'md':216 'mean':238 'mis':262 'mis-cit':261 'mismatch':37 'mode':72 'note':126 'one':233 'openjudg':66,117 'option':192,321 'output':189,213,215 'paper':257,291,316 'paperreviewpipelin':67 'pars':74 'path':129,179,198,218 'pdf':325 'pip':113 'pipelin':320 'prerequisit':111 'public':303 'py':116 'py-openjudg':115 'python':151,165,180 'queri':84 'quick':143 'rate':141,163,210 'real':61,256,290 'recommend':212 'refer':12,26,91,229 'references.bib':157,171,186 'relev':191 'report':24,101,175,217,221 'requir':125,202 'resolv':312 'resourc':318 'result':227 'review':154,168,183,326 'right':315 'run':123 'save':174 'shown':279 'skill':52 'skill-bib-verify' 'source-agentscope-ai' 'standalon':148,204 'start':144 'status':236,237 'suspect':29,107,248,281 'three':235 'titl':39,95,249,283,287 'topic-agent' 'topic-agent-skills' 'topic-ai-agent' 'topic-alignment' 'topic-evaluation' 'topic-grader' 'topic-llm' 'topic-reward' 'topic-reward-model' 'topic-rlhf' 'topic-skill-md' 'topic-skills' 'treat':271 'use':43,64 'user':46,121 'verif':51,205,328 'verifi':3,4,28,106,134,146,239 'want':47 'whether':285,294,301,309 'year':41,97,299,304 'yes':130 'your@email.com':173,188 'zh':225","prices":[{"id":"d9906930-654b-41a3-aa36-e1f56cd2c0ed","listingId":"6d80d634-a19c-4271-8029-416e8e8c47c7","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"agentscope-ai","category":"OpenJudge","install_from":"skills.sh"},"createdAt":"2026-04-18T21:57:25.913Z"}],"sources":[{"listingId":"6d80d634-a19c-4271-8029-416e8e8c47c7","source":"github","sourceId":"agentscope-ai/OpenJudge/bib-verify","sourceUrl":"https://github.com/agentscope-ai/OpenJudge/tree/main/skills/bib-verify","isPrimary":false,"firstSeenAt":"2026-04-18T21:57:25.913Z","lastSeenAt":"2026-05-02T18:53:08.010Z"}],"details":{"listingId":"6d80d634-a19c-4271-8029-416e8e8c47c7","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"agentscope-ai","slug":"bib-verify","github":{"repo":"agentscope-ai/OpenJudge","stars":585,"topics":["agent","agent-skills","ai-agent","alignment","evaluation","grader","llm","reward","reward-model","rlhf","skill-md","skills"],"license":"apache-2.0","html_url":"https://github.com/agentscope-ai/OpenJudge","pushed_at":"2026-04-30T08:18:46Z","description":"OpenJudge: A Unified Framework for Holistic Evaluation and Quality Rewards","skill_md_sha":"d64576a6efefa5eeb4cd628f45e3db2beaf53a69","skill_md_path":"skills/bib-verify/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/agentscope-ai/OpenJudge/tree/main/skills/bib-verify"},"layout":"multi","source":"github","category":"OpenJudge","frontmatter":{"name":"bib-verify","description":"Verify a BibTeX file for hallucinated or fabricated references by cross-checking every entry against CrossRef, arXiv, and DBLP. Reports each reference as verified, suspect, or not found, with field-level mismatch details (title, authors, year, DOI). Use when the user wants to check a .bib file for fake citations, validate references in a paper, or audit bibliography entries for accuracy."},"skills_sh_url":"https://skills.sh/agentscope-ai/OpenJudge/bib-verify"},"updatedAt":"2026-05-02T18:53:08.010Z"}}