An MCP control plane over the Fortran CASINO quantum Monte Carlo code
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":"casino-mcp","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.
# casino-mcp [](https://pypi.org/project/casino-mcp/) [](https://pypi.org/project/casino-mcp/) [](https://github.com/Konjkov/casino-mcp/actions/workflows/ci.yml) [](LICENSE) [](https://glama.ai/mcp/servers/Konjkov/casino-mcp) An MCP control plane over the Fortran [CASINO](https://vallico.net/casinoqmc/) quantum Monte Carlo code: write the `input` for the next calculation — and the blank Jastrow factor, backflow function and geminal wave function for the first one — start it, know what is running, stop it, and read the result as structured data instead of shipping 4000 lines of text into a model's context, including from a DMC run that is still going, which has no energy in `out` at all until its last block. > **Beta (0.6.0).** Everything below is tested against a real CASINO: the recipes against > `runqmc`'s own input check, and every file the wave function writer produces against a CASINO > test run. Interfaces may still move before 1.0. ## What it is, and what it is not CASINO already has the primitives — `opt_plan`, `runqmc --auto-continue`, `multirun`, `envmc`, `make_E_v_dt`. What it has no place for is the layer between them: machine-readable results, a memory of what was run, and the judgement calls that sit between the steps ("has the variance plateaued?", "is this timestep still in the linear regime?"). That layer is what this package is, and three rules keep it honest: 1. **No number is produced by the model.** Every value a tool returns is read from a file and carries the line it came from. What CASINO did not print comes back as `null` with a reason, never a guess. 2. **Every result is reproducible from its…
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.