OpenAgentSkill Registry Manifest
Skill: AgentBench
Slug: thudm-agentbench
Category: agent-frameworks
Description: A Comprehensive Benchmark to Evaluate LLMs as Agents (ICLR'24)

Agent fit:
- Decision: 100/100 Production-ready
- Primary fit: Coding agents
- Role: Primary pick

Supply profile:
- Track: Coding and developer agents
- Scenario: Coding agents
- Applicable agents: Claude Code, OpenAI Agents, CLI, Codex, Cursor
- Maintenance: 6mo since push
- Risk: Safe to try

Trust:
- Trust score: 87/100 Production candidate
- Audit: 89/100 Safe to try

Attribution:
- Status: Community indexed
- Source: GitHub star discovery
- Creator: THUDM
- Claim URL: https://www.openagentskill.com/skills/thudm-agentbench#claim-this-skill

Install:
npx skills add THUDM/AgentBench

URLs:
- Web: https://www.openagentskill.com/skills/thudm-agentbench
- API: https://www.openagentskill.com/api/agent/skills/thudm-agentbench
- Install API: https://www.openagentskill.com/api/skills/thudm-agentbench/install
- Repository: https://github.com/THUDM/AgentBench