{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_h7jbd4r2svwp","handle":"iflow-mcp-ziqiao-git-toolgym","url":"https://wellknown.network/agents/iflow-mcp-ziqiao-git-toolgym","links":{"self":"https://wellknown.network/agents/iflow-mcp-ziqiao-git-toolgym/record.json","html":"https://wellknown.network/agents/iflow-mcp-ziqiao-git-toolgym","markdown":"https://wellknown.network/agents/iflow-mcp-ziqiao-git-toolgym/record.md","api":"https://wellknown.network/api/v1/agents/iflow-mcp-ziqiao-git-toolgym","status":"https://wellknown.network/api/v1/agents/iflow-mcp-ziqiao-git-toolgym/status","claim":"https://wellknown.network/agents/iflow-mcp-ziqiao-git-toolgym/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/iflow-mcp-ziqiao-git-toolgym/claim.json","badge":"https://wellknown.network/agents/iflow-mcp-ziqiao-git-toolgym/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/iflow-mcp-ziqiao-git-toolgym/history","tools":"https://wellknown.network/api/v1/agents/iflow-mcp-ziqiao-git-toolgym/tools"},"ard":{"identifier":"urn:air::server:iflow-mcp-ziqiao-git-toolgym","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"iflow-mcp_ziqiao-git-toolgym","summary":"ToolGym: An Open-world Tool-using Environment for LLM Agent Evaluation","description":"# ToolGym\n\n**An Open-world Tool-using Environment for Scalable Agent Testing**\n\n[![Paper](https://img.shields.io/badge/Paper-ACL%202025-blue)](https://arxiv.org/abs/xxxx.xxxxx)\n[![Dataset](https://img.shields.io/badge/Dataset-HuggingFace-yellow)](https://huggingface.co/ToolGym)\n[![Website](https://img.shields.io/badge/Website-GitHub%20Pages-green)](https://ziqiao-git.github.io/ToolGym/)\n\n## Overview\n\nToolGym is a large-scale, open-world benchmark for evaluating LLM agents' tool-using capabilities. Built on **5,571 real tools** across **204 applications**, ToolGym enables realistic testing with:\n\n- **Long-horizon workflows**: Multi-step tasks requiring complex tool coordination\n- **Wild constraints**: Natural language requirements that must be satisfied\n- **Robustness testing**: State Controller for systematic perturbation testing\n\n## Key Statistics\n\n| Metric | Value |\n|--------|-------|\n| Total Tools | 5,571 |\n| Applications | 204 |\n| Task Instances | 3,091 |\n| Avg. Tools per Task | 4.77 |\n| Avg. Steps per Task | 7.46 |\n\n## Architecture\n\n```\n┌─────────────────────────────────────────────────────────────────┐\n│                         ToolGym                                  │\n├─────────────────────────────────────────────────────────────────┤\n│                                                                  │\n│  ┌─────────────────┐    ┌─────────────────┐    ┌──────────────┐ │\n│  │ Task Creation   │    │ Tool Retrieval  │    │    State     │ │\n│  │    Engine       │    │     Index       │    │  Controller  │ │\n│  │                 │    │                 │    │              │ │\n│  │ • Workflow      │    │ • BGE-M3        │    │ • Tool-level │ │\n│  │   Synthesis     │    │ • FAISS         │    │ • State-level│ │\n│  │ • Constraint    │    │ • 5,571 tools   │    │ • Constraint │ │\n│  │   Generation    │    │                 │    │   -level     │ │\n│  └─────────────────┘    └─────────────────┘    └──────────────┘ │\n│                                                      …","publisher":null,"homepage":"https://github.com/Ziqiao-git/ToolGym","repository":"https://github.com/Ziqiao-git/ToolGym","version":"0.1.0","license":null,"protocols":["mcp"],"tags":["mcp"],"pricing":null,"endpoints":[{"url":"pypi:iflow-mcp_ziqiao-git-toolgym","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi"}},"derived":{"capabilities":[{"slug":"dev.version-control","name":"Version Control","confidence":1,"provenance":"derived"},{"slug":"ai.model-access","name":"Model Access","confidence":0.905,"provenance":"derived"},{"slug":"ai.evaluation","name":"Evaluation & Benchmarks","confidence":0.859,"provenance":"derived"}],"categories":["ai","dev"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"iflow-mcp_ziqiao-git-toolgym","registry":"pypi","observedAt":"2026-09-15T20:21:43.571Z","publishedAt":"2026-02-26T17:32:13.627450Z","latestVersion":"0.1.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"iflow-mcp_ziqiao-git-toolgym","url":"https://pypi.org/project/iflow-mcp_ziqiao-git-toolgym/","firstSeenAt":"2026-09-09T21:25:25.448Z","fetchedAt":"2026-09-15T20:21:09.209Z","normalizedAt":"2026-09-15T20:21:09.209Z"}]},"firstSeenAt":"2026-09-09T21:25:25.448Z","updatedAt":"2026-09-15T20:21:43.571Z"}