The High-Performance Local Runtime for Autonomous Agents
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"mcp-agent-runtime","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# MCPRuntime  [](https://opensource.org/licenses/MIT) [](https://github.com/TJKlein/mcpruntime/actions/workflows/tests.yml) [](pyproject.toml) [](https://www.python.org/downloads/) [](Dockerfile) [](https://github.com/psf/black) [](benchmarks/) [](docs/benchmark_guide.md#citation) **A minimal runtime for code-first AI agents. Ships with PTC-Bench — a benchmark for comparing code-first (PTC) vs JSON-first (Function Calling) tool use.** > Build agents that generate and execute code safely. Includes 89-task benchmark to measure when code-first beats JSON-first. **Why MCPRuntime:** - ⚡ **Pluggable execution** — subprocess, Docker, or OpenSandbox (swap backends in one line) - 🧩 **Extensible patterns** — from recursive context handling to self-growing tool libraries - 📊 **Built-in benchmark** — measure PTC vs FC with your LLM provider in 60 seconds MCPRuntime decouples the **execution runtime** from the agent's reasoning loop. It provides a stable, high-performance primitive for building durable agent systems that can read, write, and execute code safely. By treating tools as importable libraries within a sandboxed environment (the **[Programmatic Tool Calling](https://www.anthropic.com/engineering/code-execution-with-mcp)** pattern), MCPRuntime enables agents to reason over large datasets and perform complex multi-step task…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.