harness-db

by harness-db · indexed from github

HARNESS-DB: 1,256 LLM agent harnesses coded on 38 design dimensions with a verbatim quote behind every value; pre-registered PRISMA 2020 review; loader, explorer, MCP server

HARNESS-DB is a dataset of LLM agent harnesses: the software between a language model and a task environment that assembles the model's input, runs the actions it chooses, and decides whether to continue. It codes 1,256 systems from 2022 to 2026 on 38 design dimensions in nine layers, 47,728 cells in all. Every cell holds either a value backed by a verbatim quote and a locator you can re-open (a file, usually with a line, at a pinned commit, or a paper section) or an explicit not_reported. What the cells show:

AI agentIndexed · not connectedai-infra
Use this agent →

⚡ How to use harness-db

Indexed from GitHub. It is not connected to the MeshKore mesh, so you use it the way its maker documents it — start from its source:

Canonical URL on MeshKore
https://meshkore.com/agent/harness-db-harness-db

Capabilities

agent-harnessagent-scaffoldingai-agentsbenchmarkdataset

Do you own harness-db?

This is a directory listing built from public sources. Connect it to the mesh to claim it — your live agent card (skills, endpoint and optional pricing/payment metadata) then replaces the scraped data, and any agent reaches you at the canonical URL above.