ToeStub MCP Server
Enables oilfield job coherence, multi-well rotation, and audit through MCP tools like toestub_job_run and toestub_job_audit.
README
ToeStub
Job OS cognitive governor for external systems — pin-sealed multi-source oilfield job coherence, explore-vs-certify firewall, multi-well rotation, partner audit, and optional MCP attachment for agent hosts (Cursor / Claude / Grok).
Pin sealed. Explore never certifies. Never λ=γ.
This is a standalone distribution extracted from the 123abc-grok research monorepo. External teams can clone, configure, and run without the protein dual-gate stack.
License: MIT
Quick start (Windows)
git clone https://github.com/RogueGringo/ToeStub.git
cd ToeStub
install.bat
menu.bat
| Script | Role |
|---|---|
install.bat |
Python deps + import check |
menu.bat |
CMD menu (cfg, smoke, rotation, audit, MCP) |
run_mcp.bat |
Start stdio MCP server for hosts |
cfg\toestub.ini |
Paths and job defaults |
Menu map
- Show configuration
- Edit
cfg\toestub.ini(Notepad) - Fixture smoke (expects certified)
- Custom run (LAS/MP/survey from cfg)
- Multi-well rotation (fixture dual well)
- Catalog runs
- Audit latest run
- MCP connect help
- Start MCP server (stdio)
T. Unit tests
I. Re-install deps - Exit
Configure
Edit cfg/toestub.ini:
[paths]
default_las = C:\path\to\well.las
default_micropulse = C:\path\to\MicroPulse_dir
default_survey = C:\path\to\SURVEY.csv
out_dir = out\job_os
[job]
channel_pack = surface_min
max_rows = 500
max_depth_mono_violations = 5
with_dynamical_topology = true
Free params are closed-set only (align_mode, window_scale, channel_pack, null_policy, survey_gate, regime_mode).
Pin knobs (depth_mono_eps, mono violations) are pin_config, not free.
MCP (agent hosts)
run_mcp.bat
Or host JSON (examples/toestub.mcp.json):
{
"mcpServers": {
"toestub": {
"command": "python",
"args": ["-m", "toestub.mcp_server"],
"env": {
"PYTHONPATH": "C:/PRIMEdEV-1/ToeStub",
"TOESTUB_REPO_ROOT": "C:/PRIMEdEV-1/ToeStub"
}
}
}
}
Point PYTHONPATH / TOESTUB_REPO_ROOT at your clone path.
Tools
| Tool | Purpose |
|---|---|
toestub_job_run |
Job OS OS-mode loop |
toestub_job_catalog |
INDEX of runs |
toestub_job_rotation |
Multi-well same pin |
toestub_job_audit |
Partner audit (no re-ingest) |
toestub_firewall_read |
Read FIREWALL.json |
CLI (operators)
set PYTHONPATH=%CD%
python job_coherence.py --os --las fixtures\mini_edr.las --out-dir out\job_os
python job_coherence.py --catalog --out-dir out\job_os
python job_coherence.py --audit out\job_os\<run_id>
python job_coherence.py --rotation examples\job_rotation_fixture.manifest.json --out-dir out\rotation
Law
- CERTIFY = pin hard seal + fixed-point (empty free board × K)
- EXPLORE = science / topology / λ1 trends —
not_acceptance - Near-miss: explore looks good, certify fails → reject for ship
- No shell tool; no soft_T / pin rewrite via free params
Design: docs/design-toestub-mcp.md
Requirements
- Windows 10+ (menu.bat) or any OS for Python/MCP
- Python 3.11+
pip install -r requirements.txt
Provenance
Kernel modules: realm/job_os, realm/dynamical_topology, realm/kb_geometry, toestub
Upstream research monorepo: RogueGringo/123abc-grok
License
Same as upstream monorepo unless otherwise noted. Contact maintainer for commercial terms.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.