MCP server that wraps the Dagster GraphQL API — manage runs, assets, schedules, sensors, and backfills from any MCP client
Nothing to run from here
ScoreCompute is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
28
tools declared
—
trust score
remote
at its publisher
—
calls last 7d
What it does
38 MCP tools: Rust computation, Discovery, composition, counterexample scans and signed checks.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user --transport http scorecompute-mcp https://scorecompute.com/mcp
# 2 — ask your agent for something these tools do
# simulate_pi, optimize_tasks, compute_orbit and 25 more
ScoreCompute costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
SCAll 1 server
scorecompute1 server
1 category
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. Nothing here runs on your machine and nothing runs on ours: this server answers at its publisher’s own address, and what you install is the connection straight to it.
Claude Codedeclared by this listing
This is a public server: it answers at its publisher’s own address and this marketplace is not in the path. Claude Code registers it in one command; --scope user makes it available in every project. The transport below is http, as the publisher's own manifest declares it.
claude mcp add --scope user --transport http scorecompute-mcp https://scorecompute.com/mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it answers at its publisher’s own address: this marketplace is not in the path, so there is no key of ours to hold and no card to add.
What it writes
One config entry pointing straight at the publisher’s own address — this marketplace is not in the path, and nothing from this listing executes on your machine.
What you approve
28 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Estimate pi with reproducible Monte Carlo sampling and a 95% Wilson interval. Select CPU or CUDA; at most 5 million points. CUDA acceleration is available for this tool only.
No parameter schema on file.
Select independent tasks exactly to maximize their total value within an integer-minute time budget. Each task can be selected once; at most 32 tasks.
No parameter schema on file.
Compute an ideal elliptical orbit around one solar mass. Time is measured from periapsis and distances are in astronomical units. This two-body model is not a real celestial ephemeris.
No parameter schema on file.
Check solar-shadow consistency from supplied image measurements: NOAA solar position, ground-plane homography from 4 to 16 control points, and Monte Carlo uncertainty. Returns compatible, contredit or intestable under the stated assumptions; never an image-authenticity verdict.
No parameter schema on file.
Check a claimed Sun or Moon position for a UTC date and location. Uses NOAA/Meeus approximations, topocentric lunar parallax, angular separation and lunar phase. The object parameter uses soleil (Sun) or lune (Moon). Returns compatible, contredit or intestable.
No parameter schema on file.
Apply necessary GRIM/GRIMMER arithmetic consistency checks to reported statistics for bounded integer observations. Examine rounded means and sample standard deviations. Passing the filters is not proof that a dataset exists; an inconsistency is not evidence of fraud.
No parameter schema on file.
Convert observed angular speed into minimum transverse speed at candidate distances using v = angular speed times distance. Compare illustrative balloon, drone, aircraft and satellite limits and check a claimed speed-distance pair. Makes no claim about object origin.
No parameter schema on file.
Compare measured shadow azimuths over time with the expected solar ephemeris. Check slope, time-lapse speed, rotation direction and individual residuals. Inputs are measured minute/azimuth pairs, not video files. Returns compatible, contredit or intestable.
No parameter schema on file.
Solve Kuhn poker with CFR+ self-play and measure exploitability using pure best responses. Compare the game value against the analytical reference -1/18. An illustrative game-theory solver, not gambling advice.
No parameter schema on file.
Compare supplied chess positions and played moves against Stockfish UCI: first-choice agreement and centipawn loss per position. Accept SAN or UCI moves. Returns a statistical profile, never a cheating verdict. The complete analysis has a 25-second deadline.
No parameter schema on file.
Screen a supplied trade series for repeated volume at nearly identical prices within a bounded time window. Report recycled-volume ratio and cycle count. This is a pattern to examine, not proof of wash trading or an accusation.
No parameter schema on file.
Test a task plan, including one produced by optimize_tasks, using independent bounded uniform duration variations. Return the probability of meeting the budget, mean duration and empirical 95th percentile. This illustrative uncertainty model is not a forecast guarantee.
No parameter schema on file.
Order 2–256 supplied points from fixed start index 0, with an optional return. Explicit planar kilometres or great-circle kilometres on a 6371.0088 km sphere. CPU Held–Karp certifies the numerical optimum only after complete search for at most 13 points; larger instances use bounded nearest-neighbour and 2-opt heuristics. Returns every leg, exact flag, stopping reason, measured time and work. Geom
No parameter schema on file.
Compute golden [-4,+6 degrees] and blue [-6,-4 degrees] photographic windows for the geometric solar centre during one strict UTC day [00:00, next 00:00), without refraction. Input must be a real Gregorian midnight from 1900 through 2100, latitude -89..89 and East-positive longitude -180..180. Return UTC millisecond windows and threshold crossings, unrounded solver durations, midpoint azimuths whe
No parameter schema on file.
Estimate a single moving object’s 3D position and velocity from geometric azimuth/elevation tracks supplied by 2–4 stationary observers. Bounded Rust CPU weighted least squares assumes constant velocity in a spherical Earth-fixed frame (R=6371.0088 km), declared angular standard deviations and supplied fixed clock offsets. Returns estimated, inconsistent, degenerate or no_convergence, component st
No parameter schema on file.
Fit original angular observations in Rust, then rank 1–64 proposed observation series by predicted position uncertainty at one common corrected-time horizon. Each candidate supplies a station, 3–120 future times, angular standard deviations and a declared cost, never candidate angles or a client-supplied baseline/covariance. Existing stations use observer_1, observer_2, etc. The local linear infor
No parameter schema on file.
Refit original angular observations together with one explicitly supplied new angular series. Rust computes the baseline itself, uses its mean only as an initialization, and includes every original/new angular pair once without adding the baseline covariance as a prior. New timestamps are already clock-corrected and must follow the baseline; existing stations retain coordinates and angular noise.
No parameter schema on file.
Search 1–32 supplied hypothetical angular-bias/clock-correction perturbations for counterexamples to a declared reconstruction-stability claim. Rust fits the original observations, applies each bounded perturbation to one observer, and refits the same spherical constant-velocity model. A counterexample can violate the position-shift limit or the requirement for a usable consistent fit; solver non-
No parameter schema on file.
Compare up to 32 predeclared combinations of deadline extensions and optional-task removals while preserving every required task. Caller supplies costs per extra minute and lost value, maximum change cost, an independent-uniform duration model, success target and fixed sample count. Search uses simultaneous Hoeffding bounds and selects the lowest-cost search-qualified examined candidate; one separ
No parameter schema on file.
Inspect the fourteen base native tool contracts, themes, units, assumptions, execution backends and composition readiness. MCP tools/list discovers all direct tools; list_composition_capabilities includes the additional complete-request bindings for triangulation and Discovery. Optional exact theme filter. The scheduling family compares task selections; approved selection_risk and route_risk chain
No parameter schema on file.
Preview a bounded robust scheduling mission without executing engines. Validate explicit duration assumptions and local tool permissions, then return up to three candidate budgets (nominal, 10% reserve, 20% reserve), bounds on engine calls and samples, or explicit blockers. Does not generate arbitrary tools or execute external code.
No parameter schema on file.
Compare up to three task selections in native Rust using optimize_tasks and, when needed, simulate_plan. Select the highest-value examined candidate whose simultaneous 95% Monte Carlo lower bound meets the requested success target under independent uniform duration uncertainty. Use an exact bound instead of simulation when possible and reuse identical selections within this mission. Return inconcl
No parameter schema on file.
Read the trusted native bank of versioned semantic contracts, engine capabilities, explicit adapters, composite capabilities and reproducible planning examples. Each individual MCP engine remains independently callable. This catalogue contains implemented capabilities only; unsupported model/unit facts explain missing adapters. No computation is executed.
No parameter schema on file.
Let native Rust compose a dependency plan from goal contracts and declared available contracts. Enforce allowed tools, locality, step and search limits. Return the selected graph, examined alternatives, blockers, search completeness and any existing-tool execution binding. Structural cost counts registered graph steps; it is not measured runtime, financial cost or scientific quality. Available fac
No parameter schema on file.
Recompile and execute an approved optimize_tasks → selected_durations.v1 → simulate_plan or route_optimizer → route_durations.v1 → simulate_plan path in Rust. Supply the engine arguments plus simulate_plan configuration (budget, uncertainty, samples, seed; omit durations). The route path also requires adapters["route_durations.v1"].speed_kmh from 1 to 150, explicit independent-uniform assumptions
No parameter schema on file.
Read the live contributor pilot: available and busy GPUs, explicitly offered CPU threads, worker supported_workloads, stale workers and verified sample throughput broken down by workload. Aggregate samples mix different calculations and are not a performance comparison. Hardware is reported, not attested. VRAM is separate per GPU, not a shared memory pool. This does not add the legacy worker inven
No parameter schema on file.
Queue a public synthetic compiled Monte Carlo job on consenting contributor clients. pi_chunk_v1 uses the fixed 31-bit integer circle model on CPU/CUDA, distinct from simulate_pi. schedule_probe_v1 is CPU-only and requires one of three compiled scenarios; each task samples a 65536-point midpoint grid with ±25% duration variation, distinct from continuous simulate_plan. No custom durations, files o
No parameter schema on file.
Read one public contributor ticket and verified progress. Completed pi jobs contain estimate; completed schedule jobs have estimate/inside null and report successes, probability and a fixed-sample Wilson95 interval under the declared discrete synthetic model. A model probability does not establish real completion. Worker-reported elapsed time, coordinator verification time and total wall time rema
No parameter schema on file.
Free
$0 forever
- Every tool — any call limit is its publisher’s, not ours
- No card, and mcprush charges nothing for it
- Runs at its publisher’s own address — this marketplace is not in the path
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install ScoreCompute or to call it, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: —
—Grade not published
No rule-by-rule report on file for this release
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
This listing is an address its publisher runs, and you connect to it directly — this marketplace is not in the path. The scanner reads packages and public source, and an address is neither. So there is no grade — not a withheld one, an unmeasured one.
Release history
This is an address its publisher runs: which release it serves is theirs to choose, so there is nothing here to pin. The history below is for reading; the current release on file is 0.17.1.
0.17.1
undated
latestNo release note was published with this version.
—
no reviews yet
Who can review
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Sign in to write one
Any signed-in account that does not publish ScoreCompute can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.