OpenAgentSkill Registry Manifest
Skill: ref-hallucination-arena
Slug: agentscope-ai-ref-hallucination-arena
Category: security
Description: Benchmark LLM reference recommendation capabilities by verifying every cited paper against Crossref, PubMed, arXiv, and DBLP. Measures hallucination rate, per-field accuracy (title/author/year/DOI), discipline breakdown, and year constraint compliance. Supports tool-augmented (ReAct + web search) mode. Use when the user asks to evaluate, benchmark, or compare models on academic reference hallucination, literature recommendation quality, or citation accuracy.

Agent fit:
- Decision: 81/100 Strong shortlist
- Primary fit: Research agents
- Role: Companion skill

Supply profile:
- Track: Research and knowledge work
- Scenario: Research agents
- Applicable agents: Claude Code, OpenAI Agents, CLI, Codex, Cursor
- Maintenance: 1mo since push
- Risk: Needs review

Trust:
- Trust score: 72/100 Strong shortlist
- Audit: 76/100 Needs review

Attribution:
- Status: Registry indexed
- Source: github fast track
- Creator: agentscope-ai
- Claim URL: https://www.openagentskill.com/skills/agentscope-ai-ref-hallucination-arena#claim-this-skill

Install:
npx skills add agentscope-ai/OpenJudge --skill ref-hallucination-arena

URLs:
- Web: https://www.openagentskill.com/skills/agentscope-ai-ref-hallucination-arena
- API: https://www.openagentskill.com/api/agent/skills/agentscope-ai-ref-hallucination-arena
- Install API: https://www.openagentskill.com/api/skills/agentscope-ai-ref-hallucination-arena/install
- Repository: https://github.com/agentscope-ai/OpenJudge/tree/main/skills/ref-hallucination-arena