MCP Connector

Connect Claude, and any MCP-compatible AI assistant, directly to the OCS interactive science tools via the Model Context Protocol.

omegacentauri.me · Live endpoint · Updated 2026-09-02

Endpoint

https://mcp.omegacentauri.me/mcp

The server is a stateless Cloudflare Worker running on the Streamable HTTP MCP transport (2025-11-25). The browser calculators run client-side. The MCP server returns deep-links and metadata, except for the seven compute kernels and run_chain, which run on the worker and receive the inputs you send them.

Connecting your AI assistant

The OCS endpoint uses the Streamable HTTP MCP transport (spec 2025-11-25) and works with any compliant client. Select your client below.

1

Open the Claude desktop app and go to Settings → Connectors.
2

Click Add connector and select Custom MCP server.
3

Paste the endpoint URL: https://mcp.omegacentauri.me/mcp
4

Name it Omega Centauri Society and save. Tools appear under Read-only connectors.

Health check: mcp.omegacentauri.me/health returns {"status":"ok"}. Verify the server is up before connecting.

Add to ~/.cursor/mcp.json (create it if absent). Cursor picks up changes on the next project reload.

{
  "mcpServers": {
    "omega-centauri-society": {
      "type": "http",
      "url": "https://mcp.omegacentauri.me/mcp"
    }
  }
}

Tip: ~/.cursor/mcp.json applies globally; a per-project override lives at .cursor/mcp.json in the workspace root.

VS Code 1.99+ native MCP — add to .vscode/mcp.json in your workspace:

{
  "servers": {
    "omega-centauri-society": {
      "type": "http",
      "url": "https://mcp.omegacentauri.me/mcp"
    }
  }
}

Cline extension — open Cline settings → MCP Servers → Edit config and add:

{
  "omega-centauri-society": {
    "type": "http",
    "url": "https://mcp.omegacentauri.me/mcp",
    "disabled": false
  }
}

Any HTTP client can call the endpoint directly. The tools/list method confirms connectivity:

curl -X POST https://mcp.omegacentauri.me/mcp \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}'

Transport: The server uses Streamable HTTP (MCP spec 2025-11-25). Long-running responses may stream as text/event-stream.

W5 — Agentic literature loop

W5 combines the OCS MCP with NASA ADS and an arXiv MCP in a single agent session. The agent fetches recent mass-constraint papers, propagates any new measurements through the OCS constraint chain, and outputs an updated evidence ledger with deeplinks.

Required MCP servers:

1 · OCS https://mcp.omegacentauri.me/mcp — no setup beyond connection
2 · NASA ADS github.com/prtc/nasa-ads-mcp — requires a free NASA ADS API key
3 · arXiv Any arXiv MCP — search the MCP registry for arxiv and follow that project's README

Three-server config skeleton (Cursor / VS Code — fill in the ADS and arXiv entries from each project's README):

{
  "mcpServers": {
    "omega-centauri-society": {
      "type": "http",
      "url": "https://mcp.omegacentauri.me/mcp"
    },
    "nasa-ads": {
      "command": "...",
      "args": ["..."],
      "env": { "ADS_API_KEY": "YOUR_KEY" }
    },
    "arxiv": {
      "command": "...",
      "args": ["..."]
    }
  }
}

Agentic loop prompt: the copy-paste prompt now lives in the Example Prompts Library as agentic-literature-loop. Connect the three servers above, open the library card, and copy the prompt from there; the library version is the maintained one.

Example prompts

Copy-paste agent prompts for common OCS research tasks now live in the Example Prompts Library: 30 prompts rendered from a machine-readable SSOT, each ending at a surface you can verify without trusting us. The recipes that used to sit on this page are superseded by these library cards:

E2 find-the-right-tool · survey the catalog, run one tool, mint a deeplink that reproduces the run
S2 run-the-evidence-window · the machine chain that turns ten published measurements into the allowed IMBH mass window, then breaks it
V1 stack-the-constraints · all ten constraint channels with floor/ceiling provenance and page-versus-worker parity
G1 agentic-literature-loop · the OCS + NASA ADS + arXiv loop (successor to the W5 block above)

The library also carries everyday tasks, by-domain briefs, education tracks, and all 42 catalog chains as reading paths; it supersedes every copy-paste recipe this page used to carry.

Available tools

The worker exposes 11 callable tools: 4 meta tools (search/deep-link/verify/execute) and 7 server-side compute kernels. Live-verified against mcp.omegacentauri.me/mcp tools/list. It also serves the Example Prompts Library over the MCP prompts capability: prompts/list returns the 30 verification prompts and prompts/get returns one in full.

Meta tools

list_ocs_tools

Search the OCS interactive calculator suite. Returns deep-links to client-side browser tools. 22 of the 35 catalog tools are prefill-enabled: the returned URL includes an #in= fragment that opens the tool pre-filled with your values.

Parameters (all optional)

query string — free-text search against title and description
category string — filter: imbh-evidence · bh-physics · fermi-paradox · fermi-seti · mth · kardashev
register string — filter: peer-reviewed · speculative
limit number — max results (default 20)

build_ocs_workflow_links

Constructs an ordered set of ready-to-use deep-links for a named OCS scenario or workflow chain, or an ad-hoc sequence of tools. Pass fields per step to receive pre-filled URLs via the #in= fragment convention. These are narrative reading paths for a human or agent to follow, not server-executed pipelines; use run_chain for a server-executed pipeline instead.

Parameters

chain string, exclusive with steps — name of a pre-defined chain (42 available, see below)
steps array, exclusive with chain — ad-hoc list of {tool, fields?} objects

verify_execution_hash

Independently verifies a ChainGraph artifact (ChainGraph Standard §4 JCS). Recomputes SHA-256 over the canonical (RFC 8785 JCS, sorted-key, whitespace-stripped) JSON of {policy_parameters, output_payload} and compares it to the claimed execution_hash. Works on artifacts from any ChainGraph-conformant source, not only OCS tools.

Parameters

artifact object, or the three fields below: a full ChainGraph artifact
policy_parameters / output_payload / claimed_hash: alternative to artifact

run_chain

Executes a ChainGraph chain server-side (OCG Standard §21), running the actual compute kernels (constraint-stacker, bayes-factor-router, jwst-accretion-ledger, gwtc-remnant-classifier, roman-microlensing) in sequence, not deep-links. Supports §21.4 decision gates: an RFC 6901 pointer into a step's output_payload is matched against ordered rules (eq/neq/gt/gte/lt/lte/in/present/absent) to route to the next step.

Parameters

chain string, exclusive with steps: one of 3 machine-executable chains: imbh-evidence-window (default) · evidence-threshold-routing · accretion-eligibility-fastfail
steps array, exclusive with chain: ad-hoc ordered list of {tool_id, fields?, gate?} objects

Compute kernels (7)

Each returns a hash-anchored OCS ChainGraph artifact JSON; each mirrors a client-side browser tool but also runs server-side for chaining via run_chain.

constraint_stacker

Computes the currently allowed IMBH mass window for NGC 5139 given active observational constraints. Mirrors omegacentauri.me/tools/constraint-stacker.html. Häberle 2024 lower bound (8,200 M☉) exceeds Bañares-Hernández 2025 upper limit (6,000 M☉), irreconcilable under default settings; the tool reports the resulting tension rather than collapsing it.

bayes_factor_router

Classifies a Bayes factor (BF or 2·lnBF) into a categorical evidence-strength token using the Jeffreys (1961) / Kass & Raftery (1995, JASA 90:773) scale: supports_null · weak · substantial · strong · decisive. Used as a §21.4 decision gate for model-selection chains.

jwst_accretion_ledger

Computes the IMBH mass above which predicted ADAF/Bondi accretion luminosity would exceed the JWST NIRCam/MIRI bolometric upper limit for NGC 5139 (Chen et al. 2025, arXiv:2511.20945), given an assumed radiative efficiency ε and ambient gas density at the Bondi radius. Model-dependent: this is a conditional accretion constraint, not a direct mass measurement.

gwtc_remnant_classifier

Classifies a compact-binary merger (GWTC-style component masses m1, m2, optional aligned spins) by chirp mass, total mass, mass ratio, remnant mass, final spin, pair-instability mass-gap membership (~60–130 M☉, model-dependent edges), and an IMBH-remnant classification token for chain routing. Uses Jiménez-Forteza et al. 2017 (PRD 95, 064024) fits.

roman_microlensing

Computes Einstein radius (θ_E, mas) and Einstein crossing timescale (t_E, days) for a gravitational microlensing event observable by the Nancy Grace Roman Space Telescope, using the Gould (2000, ApJ 542, 785) formalism, plus a yield-scaled stellar event estimate from Penny et al. (2019, arXiv:1808.02490).

apophis_flyby_geometry

Computes close-approach geometry for the 2029-04-13 flyby of asteroid 99942 Apophis (perigee ~38,017 km from Earth's centre): tidal acceleration differential across the asteroid, perigee speed via vis-viva, and GEO-ring crossing status.

rubin_alert_throughput

Estimates nightly transient-alert count and data-rate budget for the Vera C. Rubin Observatory LSST. Default parameters from Ivezić et al. (2019, ApJ 873, 111): ~1,000 visits/night, ~10,000 alerts/visit.

Named chains (42)

Pass any of these as the chain parameter to build_ocs_workflow_links.

scenario-breaking-degeneracy scenario-imbh-evidence scenario-fermi-honest scenario-fermi-mth-crossover scenario-dark-cluster-alt scenario-bh-physics scenario-bh-tap scenario-decision-tree scenario-fermi-five-ways scenario-dvali-signal scenario-dwarf-inheritance scenario-every-way-it-eats scenario-kardashev-two scenario-lisas-shot scenario-oc-vs-gc scenario-proposal-bridge scenario-reconnection-bz scenario-spin-up-economics scenario-spin-up-tap scenario-wait-compute scenario-cosmology-browser workflow-mth-compute-budget workflow-fermi-mth-crossover workflow-bh-energy-budget workflow-wait-or-act workflow-constraint-window workflow-survivability workflow-spin-economics workflow-evodevouniverse workflow-tde-detection workflow-grabby-fermi workflow-mass-tension workflow-naked-imbh workflow-accretion-to-spin workflow-multimessenger-campaign workflow-cns-reproduction workflow-settle-imbh-question workflow-black-hole-101 workflow-kardashev-ladder scenario-paper-e-feasibility scenario-migrate-or-stay scenario-background-free

Prefill-enabled tools

22 of the 35 catalog tools accept #in=<base64url(JSON)> URL fragments (verified against OCS_APPLY_PREFILL usage in the tool source, matching tools-manifest.json's _meta field). Pass values as a fields object in build_ocs_workflow_links steps to receive a pre-filled URL.

Log10 param warning: radio-seti — all 6 params (eirp, dist, aeff, tsys, tau, bw) are log₁₀. qpo-mass-spin — lognu is log₁₀(Hz). bz-kardashev — hash keys are spin (not a) and power (not P). Do not pass linear values to log₁₀ parameters.

constraint-stacker velocity-dispersion dark-cluster astrometric-microlensing qpo-mass-spin drake-monte-carlo great-filter radio-seti bekenstein-landauer bz-kardashev evidence-ledger detection-forecast observing-campaign-planner bh-scale-comparator infall-survival shadow-imaging kardashev-meter energy-translator scifi-tech-auditor flyby-survival flyby-survival-simulator imbh-fuel-budget

Technical details

Transport: Streamable HTTP, MCP spec 2025-11-25. POST /mcp, JSON-RPC 2.0, stateless (no session IDs).

Runtime: Cloudflare Workers (free plan), deployed globally. Data fixture (tools-manifest.json) served via ASSETS binding, cached per isolate.

CORS: omegacentauri.me, claude.ai, app.claude.ai, and localhost are in the allowed-origins set.

Privacy: The browser calculators run entirely client-side. Through MCP, the search, deep-link and hash-verification tools return only URLs and metadata. The seven compute kernels (constraint_stacker, bayes_factor_router, jwst_accretion_ledger, gwtc_remnant_classifier, roman_microlensing, apophis_flyby_geometry, rubin_alert_throughput) and run_chain compute on the worker, so the inputs you pass to them are sent to the worker and the results come back from it. The worker's logging is limited to privacy-scoped counters (no raw query text or IP addresses).

Source: github.com/PostOakLabs/OCS