How is this different from connecting Prisma MCP directly?

Connected directly, an agent sees every tool the server exposes and holds the credential itself. Through TrustGate you choose which tools each agent gets, and every call is logged against the agent that made it.

What if the Prisma MCP API changes?

We track the Prisma MCP schema, so the tool catalogue stays current. Tools you have already allowed keep working, and anything new stays denied until you enable it.

Can I give different agents different Prisma tools?

Yes. Access is set per agent: one can get a handful of read tools, another the full set. You grant the whole MCP or single tools within it.

Where can I see what my agents did in Prisma?

Every call is logged against the agent that made it, with the tool it used. Activity can be sent on to your own monitoring.

Do I need to run the Prisma MCP server myself?

No. TrustGate connects to the Prisma MCP server and sits in front of it. Your agent only talks to your TrustGate URL, never to Prisma directly.

Connect Prisma MCP to your AI agents, securely.

Q: Is Prisma MCP secure?

The server itself exposes 31 tools, and some of them can write to your Prisma account. Behind TrustGate an agent only sees the tools you allow, requests and responses are inspected, and every call is logged. Anything outside the allowlist is denied before it reaches Prisma.

One secured endpoint for every Prisma call your AI agents make, scoped per user, inspected, and logged.

Get started for free

Get a demo

Try Prisma MCP now

What your agents can ask once Prisma MCP is connected.

Select the tools your AI agents can call

31 tools from the Prisma MCP schema.

Showing 31 tools

fetch_workspace_detailsreadFetches the details of a Prisma Postgres workspace.
list_prisma_postgres_databasesreadFetches the list of available Prisma Postgres databases for the user's workspace.
create_prisma_postgres_databasewriteCreates a managed Prisma Postgres database and a new project.
delete_prisma_postgres_databasehigh riskPermanently deletes a Prisma Postgres database by ID. Cannot delete a project's default database.
list_prisma_postgres_connection_stringsreadFetches the list of available connection strings for a given database ID.
create_prisma_postgres_connection_stringhigh riskCreates a new connection string for a Prisma Postgres database, issuing a live database credential.
delete_prisma_postgres_connection_stringwriteDeletes a connection string by ID.
list_prisma_postgres_backupsreadLists the available automated backups for a Prisma Postgres database.
create_prisma_postgres_recoverywriteRestores a Prisma Postgres database to a new database from a given backup ID.
introspect_database_schemareadReads the schema of a Prisma Postgres database to show its structure.
execute_sql_queryhigh riskRuns SQL to read, modify or delete data in a Prisma Postgres database, without changing its schema.
execute_prisma_postgres_schema_updatehigh riskExecutes a schema update on a Prisma Postgres database.
list_object_store_bucketsreadLists the object-store buckets in the workspace, optionally filtered by project ID.
create_object_store_bucketwriteCreates a new object-store bucket in a project.
delete_object_store_buckethigh riskPermanently deletes a bucket, every object stored in it, and all of its access keys.
create_object_store_bucket_keyhigh riskCreates an S3-compatible access key for a bucket.
delete_object_store_bucket_keywriteDeletes a bucket access key. The key stops working immediately.
list_prisma_compute_appsreadLists the Prisma Compute apps in the selected workspace, optionally filtered by project.
list_prisma_compute_buildsreadLists builds newest first, with optional project, branch and state filters.
list_prisma_compute_deploymentsreadLists the deployments for a Compute app, newest first.
get_prisma_compute_deployment_logsreadReads the runtime logs for a deployment, one page at a time.
promote_prisma_compute_deploymenthigh riskMakes a running deployment the live version of an app.
rollback_prisma_compute_apphigh riskMakes an existing deployment live, starting it first if it is stopped.
start_prisma_compute_deploymentwriteStarts an existing deployment from an already uploaded build. Does not build or upload the app.
stop_prisma_compute_deploymenthigh riskStops a deployment, disconnecting it from the app's public URL if it is the live version.
delete_prisma_compute_deploymenthigh riskPermanently deletes a deployment, stopping it first if needed.
delete_prisma_compute_apphigh riskPermanently deletes an app. Setting deleteDeployments to true also stops and deletes its deployments; otherwise active deployments block the deletion.
set_prisma_compute_env_varwriteSets an environment variable for a project's production or preview environment, or overrides it for one preview branch. Values are encrypted and never returned.
delete_prisma_compute_env_varwritePermanently deletes an environment variable by ID.
get_prisma_composer_topologyreadShows the saved connections between a project's Composer services and resources, with Console links and stored app URLs. The result may be out of date and does not check whether apps are running.
search_prisma_documentationreadAnswers a natural-language question using the official Prisma documentation, returning a cited answer with links back to the docs.

What Prisma MCP exposes

Prisma MCP groups its 31 tools into 10 categories, each with a suggested starting policy.

All 31 tools as JSON, ready to paste.

Category	Covers	Access	Suggested policy
Compute Deployment Management	Promoting, rolling back, starting, stopping and deleting deployments and apps	Read / Write	These tools decide which version is live and can disconnect an app's public URL; approve every one
Object Storage	Buckets and S3-compatible access keys	Read / Write	Deleting a bucket removes its objects and all its access keys; a deleted key stops working immediately
Workspace & Databases	Workspace details, database listing, creation and deletion	Read / Write	Deletion is permanent and cannot touch a project's default database
Compute Apps & Builds	Apps, builds, deployments and deployment runtime logs	Read	Read-only; logs are paged one at a time
Connection Strings	Listing, creating and deleting the connection strings for a database	Read / Write	A connection string is a live database credential; treat creation as credential issuance
Schema & Queries	Schema introspection, SQL execution and schema updates	Read / Write	execute_sql_query changes data but not schema; schema updates are a separate tool
Backups & Recovery	Listing automated backups and restoring from one	Read / Write	Recovery restores into a new database rather than overwriting the original
Compute Environment Variables	Setting and deleting environment variables per environment or preview branch	Read / Write	Values are encrypted and never returned, so an agent cannot read back what it set
Composer Projects	The saved connections between a project's Composer services and resources	Read	The topology may be out of date and does not check whether apps are running
Documentation	Natural-language questions answered from the official Prisma documentation with citations	Read	Allow documentation lookups

Connect Prisma MCP to your AI agent

Four steps, under ten minutes.

01
Sign in to NeuralTrust
Open the NeuralTrust platform, sign in and go to Getting started, then Connect your tools.
02
Choose your agent
When asked which agent you use, pick Claude Code, Codex, Cursor, Windsurf, Copilot or your own code.
03
Connect Prisma
Find Prisma in the tool list and connect it.
04
Connect your agent
Follow the steps on screen to add the config to your agent, then send the test prompt.

Permissions down to the single tool

You do not allow Prisma MCP or block it. You allow the exact tools each agent can call, one by one.

Prisma MCP	31 TOOLS AVAILABLE
Code review agent	4 tools, read only
Issue triage bot	9 tools
Platform team assistant	all 31 tools

Tool by toolConnecting the server is not the unit of access. Each tool is granted on its own.
Default denyA tool you have not enabled is never exposed to the agent.
Approval on writesTools that change or delete things can require a human to confirm.

Is Prisma MCP secure?

One endpoint instead of raw credentials. Every call is checked against policy and logged.

AI agentsConnect through your MCP client configuration.	NeuralTrustApply controls based on your configuration.	Prisma MCP31 documented tools.
Coding assistants	Connection & credentialsper user	Compute Deployment Management6 tools
Internal copilots	Tool access policiesdeny by default	Object Storage5 tools
Workflow automations	Request & response inspectioninline	Workspace & Databases4 tools
Compatible MCP clients	Activity monitoringalerts and logs	All other tools16 tools