Chroma’s official MCP server: vector search and collection management for RAG workflows.
Schemity 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.
What it does
Local database MCP server: the agent reads your schema and stages edits, never holding credentials.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user schemity-mcp '--' npx -y schemity-mcp
# 2 — ask your agent for something these tools do
# list_diagrams, get_schema, get_context_views and 16 more
Schemity costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts Schemity as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
This is a public server: you run it yourself and this marketplace is not in the path. You need Node.js — the current LTS (24) is the safe choice; npx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user schemity-mcp '--' npx -y schemity-mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
19 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.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
Diagrams open in Schemity. Call first; every other tool takes a diagram id from here.
Takes no parameters.
Entities, fields, relations and legends (with their entities) of a diagram, or of one context view. No coordinates. Absent keys mean default: not null, not unique, plain field. Per entity, when present: uniques (multi-column), checks and indexes, each with its name. An index on an expression,…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
| context_view | string | no | Context view name; omit for the whole diagram |
Context views of a diagram with their entities, the dependencies between views, and cycles. Each dependency lists its relations, which of them are virtual (not declared as foreign keys in the database), and its kind: fk, virtual or mixed.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
Relations into and out of one entity, with the columns on both sides.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
| entity | string | yes | Entity (table) name |
Markdown data dictionary of a diagram or one context view. Larger than get_schema; ask for it only when descriptions matter.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
| context_view | string | no | Context view name; omit for the whole diagram |
Unsaved edits of a diagram, as the undo history that produced them (newest last, at most 30). A step that moved or reshaped more than 20 tables or lines counts them per scope in movedOrReshaped instead of listing each. For what a migration would cost, or its SQL, call analyze_impact.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
What the pending migration of a connected diagram would cost: data loss, failing statements, rewrites, dependents. Findings only, unless include_sql is true.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
| include_sql | boolean | no | Also return the migration's statements, in order. Only when the user wants the SQL. Schemity never runs it for you |
Same analysis for a hand-written SQL migration against a connected diagram's database. Nothing is run.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
| sql | string | yes | The migration SQL |
Schema findings for a diagram: conventions, missing indexes, type mismatches. Facts, not verdicts. Over 100 findings, each rule lists its first 10 and counts has every rule's total; pass rule for all of one.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
| rule | string | no | Only this rule's findings, all of them, e.g. fk-no-covering-index |
Relation lines in the live layout (unsaved edits included). overlaps: pairs where a portion of one line lies on the other or within 1px of it (reads as one line), with the shared length in px. crossings: how many times two lines cross. bends: how many corners the lines turn through in all.…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
| context_view | string | no | Context view name; omit for the whole diagram |
Edit the active main diagram: schema, legends, context views. Edits land as unsaved changes the user reviews in History, then saves or undoes; nothing is written to the database. All or nothing. Deleting an entity, field, relation, legend or context view asks the user first and may take minutes,…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
| actions | array | yes | — |
Exact row count for a finding's probe (full table scan: only when the user asks). Pass the probe object from an analyze_impact or analyze_migration_file finding, or {kind:'rows'} for the whole table.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
| table | string | yes | — |
| probe | object | yes | A finding's probe |
Group entities of the active main diagram into legends. Decide groups from the source code when you have it, else from the user's description, else from table names. Schemity sizes each legend and places it beside the groups it shares relations with (the most connected first, whatever order you…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
| groups | array | yes | — |
| full_size | boolean | no | Also fit each grouped entity's height to show all its fields, like its own fit button. Grouped entities always take the width that shows their names and types in full; default false keeps the height t |
| relayout | boolean | no | Place existing legends afresh by their relations too. Default false keeps them where they are. |
| beside | object | no | Put the legends this call places afresh on one side of an existing legend that is not in groups, e.g. {legend: "Inventory", side: "right"}: the first next to it, the rest beside each other, none cross |
Place entities of the active main diagram in rows, left to right and top to bottom, columns lined up: for a few tables that should sit a certain way round (group_entities is for domains). Schemity spaces them from the corner they occupied and routes relations; no coordinates. Entities not listed…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
| rows | array | yes | Rows of entity names, e.g. [["users", "orders"], ["items"]] |
Re-route every relation of the active main diagram around the entities so lines do not cross tables, go around each other where that is cheap, and share corridors as parallel lanes. Deterministic: unchanged entities give the same routes (status unchanged, no undo step). Check the result with…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
Open the Schemity app, or bring it to the front. Returns the open diagrams, or a message when none is open.
Takes no parameters.
Every saved diagram, open or not, best match first. The hint matches the diagram name and its database alike (users rarely know which they remember), then the workspace. Show the user the results and let them pick, then open_diagram.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| query | string | no | Name hint from the user; empty lists all |
| limit | integer | no | At most this many, default 10 |
Bring a diagram to the screen, opening it if it is not open; other tabs stay open. Opening one with a database may ask the user for their OS password (keychain) or to approve a password command, so tell them first.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from find_diagrams or list_diagrams |
Draw a throwaway picture in Schemity for the user to look at. One of: pending, the diagram's unsaved edits as the migration they make (changed tables are marked and every change is listed, constraints and indexes included; not after your own propose_changes, whose edits the user already sees on…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| diagram | string | yes | Id from list_diagrams |
| pending | boolean | no | Picture the unsaved edits: against the database, or with none, against the oldest state in History |
| sql | string | no | A migration file's SQL. Needs a connected diagram |
| entities | array | no | Entity (table) names to show as they are |
| title | string | no | Short caption for the picture, for example the migration's purpose |
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under MIT
- Runs on your machine — nothing of it reaches our gateway
- 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 Schemity or to call it, at any volume.
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.
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | high | capability blast radius (high) — client exposure if the model is manipulated | −6 |
| verification-discount | repo | publisher verification (public source) — no provenance, but the source is public and inspectable | −1 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
No tools were enumerated, so prompt-injection, capability and toxic-flow analysis had no tool surface to inspect. To grade a package’s real runtime tools, scan the running server: --command "npx -y <package>".
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. high is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| No per-tool rows for this release — see the note below. | ||
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.
Supply chain and provenance
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Always latest — the install command names no version, so it installs the newest release when it runs; the current release on file is 1.0.0. Pin a release below and the command asks for exactly that one.
No release note was published with this version.
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.
Any signed-in account that does not publish Schemity can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
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.