Show me all users who signed up this week and their latest query counts.
S
prismamcp_execute_query
85ms
Prisma agent
14 users signed up Mon–Sun. Top query load: alice@acme.com (1,240 queries), bob@globex.com (987 queries), carol@initech.com (734 queries).
Sources: users table, query_logs table, Jun 15–21
prismamcp
2 tables
18:29
Message Claude...
Trusted by teams shipping agents to production
Tools your database agent reaches for on Prisma MCP, scoped per user.
CALL ANY TOOL
Each tool maps to a Prisma operation your agent can call against any connected Prisma Postgres instance.
prismamcp_create_object_store_bucket
Create object store bucket
Create a new object-store bucket in the given project. On success, use the returned bucket id to generate access credentials.
Parameters
Name
Type
Required
Description
projectId
string
Required
The project ID to create the bucket in (e.g. proj_xxx).
branchId
string
Optional
Branch ID (e.g. br_xxx) to scope the bucket to. Defaults to the project default branch.
name
string
Optional
A display name for the bucket. Generated if omitted.
prismamcp_create_object_store_bucket_key
Create object store bucket key
Create an S3-compatible access key for an object-store bucket. The secret access key is returned exactly once and never stored, so it must be saved immediately.
Parameters
Name
Type
Required
Description
bucketId
string
Required
The bucket ID to create the key for (e.g. bkt_xxx). Use the List Buckets tool to find it.
role
string
Required
Access level: 'read' for read-only, 'read_write' for full access.
name
string
Optional
A display name for the key. Generated if omitted.
prismamcp_create_prisma_postgres_backup
Create prisma postgres backup
Create an automated backup for a Prisma Postgres database. Note: on-demand backup creation is not currently supported; backups are created automatically by the system.
Parameters
Name
Type
Required
Description
databaseId
string
Required
The unique identifier of the Prisma Postgres database. Get it from List Databases.
projectId
string
Required
The unique identifier of the Prisma project. Get it from List Databases.
Permanently delete a connection string by its ID. This action cannot be undone.
Parameters
Name
Type
Required
Description
id
string
Required
The unique identifier of the connection string. Get it from List Connection Strings.
prismamcp_delete_prisma_postgres_database
Delete prisma postgres database
Permanently delete a Prisma Postgres database by its ID. This action cannot be undone.
Parameters
Name
Type
Required
Description
databaseId
string
Required
The unique identifier of the Prisma Postgres database. Get it from List Databases.
prismamcp_execute_prisma_postgres_schema_update
Update execute prisma postgres schema
Execute a DDL schema update on a Prisma Postgres database. Use for schema changes only; use Execute SQL Query for data reads and writes.
Parameters
Name
Type
Required
Description
databaseId
string
Required
The unique identifier of the Prisma Postgres database. Get it from List Databases.
projectId
string
Required
The unique identifier of the Prisma project. Get it from List Databases.
query
string
Required
The SQL query to execute against the database.
prismamcp_execute_sql_query
Query execute sql
Execute a SQL query on a Prisma Postgres database and return the results as JSON. Does not have permission to run schema updates.
Parameters
Name
Type
Required
Description
databaseId
string
Required
The unique identifier of the Prisma Postgres database. Get it from List Databases.
projectId
string
Required
The unique identifier of the Prisma project. Get it from List Databases.
query
string
Required
The SQL query to execute against the database.
prismamcp_fetch_workspace_details
Fetch workspace details
Retrieve details of the current Prisma Postgres workspace, including plan limits and usage.
Parameters
Name
Type
Required
Description
—
No parameters required
prismamcp_introspect_database_schema
Introspect database schema
Introspect and return the schema of a Prisma Postgres database as JSON.
Parameters
Name
Type
Required
Description
databaseId
string
Required
The unique identifier of the Prisma Postgres database. Get it from List Databases.
projectId
string
Required
The unique identifier of the Prisma project. Get it from List Databases.
prismamcp_list_object_store_buckets
List object store buckets
List object-store buckets in the workspace, 100 per page, optionally filtered by project ID. Use the returned id as bucketId in other bucket tools.
Parameters
Name
Type
Required
Description
cursor
string
Optional
Pagination cursor from a previous response's nextCursor.
projectId
string
Optional
Filter buckets by project ID (e.g. proj_xxx). Omit to list all buckets.
prismamcp_list_prisma_postgres_backups
List prisma postgres backups
List all available automated backups for a Prisma Postgres database.
Parameters
Name
Type
Required
Description
databaseId
string
Required
The unique identifier of the Prisma Postgres database. Get it from List Databases.
prismamcp_list_prisma_postgres_connection_strings
List prisma postgres connection strings
List all connection strings for a Prisma Postgres database.
Parameters
Name
Type
Required
Description
databaseId
string
Required
The unique identifier of the Prisma Postgres database. Get it from List Databases.
prismamcp_list_prisma_postgres_databases
List prisma postgres databases
List all Prisma Postgres databases in the workspace. Use the returned id as databaseId in other tools.
Parameters
Name
Type
Required
Description
—
No parameters required
prismamcp_search_prisma_documentation
Search prisma documentation
Search Prisma's official documentation and knowledge sources to answer questions about Prisma Postgres, Prisma ORM, Accelerate, Optimize, schema design, and migrations. Returns an answer grounded in the docs, with citations.
Parameters
Name
Type
Required
Description
query
string
Required
Natural-language question about Prisma: Prisma Postgres, Prisma ORM, Accelerate, Optimize, schema design, migrations, etc.
Build your Agent
Same per-user vault and scope enforcement regardless of which AI framework your agent runs on.
Show me the migration history for the past 30 days.
SEE HOW AUTH WORKS
The developer authorizes once. Every subsequent agent call resolves their vaulted API key with scope enforcement before touching the database.
1
Authorize
Your user connects
Prisma MCP
once. We tie it to their identity and the meetings they approved — no shared bot account, no org-wide access
Who:
user ‘A’
when:
Once per user
access:
Limited to user
2
Store
Their
Prisma MCP
token lives in a vault scoped to them. User A's meetings are never reachable by an agent acting for user B, even on the same connection
vault:
encrypted
scope:
per-user
tokens:
auto-refreshed
3
Resolve
When your agent calls a
Prisma MCP
tool, we fetch the right token server-side. It never touches your agent, never appears in the LLM context, never shows up in your logs
speed:
~40ms
check:
before every call
seen by:
nobody
4
Audit
Every
Prisma MCP
tool call is logged — who triggered it, which meeting was fetched, what came back. 90 days of history, tied to the user who authorized it
history:
90 days
export:
SIEM-ready
logged:
every call
Test other agents
See the same per-user auth pattern enforced across every other developer tools connector.
Engineering Teams
DevOps assistant agent
Polls GitHub for failing checks and stale PRs, opens Linear issues for the ones that need work, and posts a daily digest to Slack. It acts as the engineer, not a shared service account.
Pulls commits from GitHub and GitLab, tracks issue movement in Jira, and posts a per-engineer standup brief to Slack. Each engineer's activity is read on their own delegated OAuth.
Reads merged GitHub PRs, groups them into structured release notes, publishes the page to Notion, and announces the release in Slack. Every call runs on the engineer's own delegated OAuth.
Polls Slack for new messages, classifies bugs and support requests with a LangGraph router, files GitHub issues or Zendesk tickets, and confirms in the thread.
Secure your agent's access. Connectors ship in minutes
01.
Shared tokens break per-user analytics
A shared database URL works in a demo. In production every query looks like a service account. Scalekit resolves the real developer credential so your audit log reflects who actually ran each query.
// shared DATABASE_URL audit → service_account
// scalekit audit → alice@acme.com ✓
02.
Authentication is not authorization
Storing an API key is table stakes. The production question: is this agent, calling this Prisma tool, for this tenant, within the permitted scope? Scalekit enforces that check on every call.
Each developer's database credentials need real isolation. Scalekit namespaces every vault entry per tenant so a misconfigured agent can never reach a neighbor's Prisma Postgres instance.
Every new connector you add is another auth flow to maintain. With Scalekit, Prisma MCP and every connector that follows inherits the same vault, refresh logic, and audit chain. Zero new auth code per connector.
// DIY: N connectors prisma_oauth.py ȕ1 github_oauth.py ȕ1 ...
“Our agents act across Salesforce, Gong, Google Drive, and more, on behalf of every customer. Scalekit behind the scenes meant we can keep adding tools without ever rebuilding how credentials or tool calling work.”
Venu Madhav Kattagoni
Head of Engineering / Von
FAQs
Frequently Asked Questions
Does every user need their own Prisma API key, or can we share one?
Every user needs their own credential. Scalekit vaults a key per user and resolves it at call time, so queries are attributed to the right developer and your audit log stays accurate. A shared key collapses all activity into one identity, which breaks both attribution and fine-grained access control.
How does Scalekit store Prisma API keys?
Keys are encrypted at rest with AES-256 and decrypted only at the moment a tool call is authorized. The plaintext credential is never written to logs, never passed through agent context, and is inaccessible outside the vault resolution path.
Can I limit which Prisma MCP tools a specific user or role can call?
Yes. Scalekit's listScopedTools API returns only the tools a given user is permitted to invoke. You pass that list to your agent at session start, so the agent cannot even attempt a tool outside the permitted scope for that user.
What happens if a user revokes their Prisma API key or I delete their vault entry?
Any subsequent tool call that tries to resolve that credential will fail with an authorization error before it reaches Prisma. The agent receives an explicit rejection rather than a stale or empty response, so your error handling can prompt the user to re-authorize.
Does Scalekit support Prisma Postgres connection strings as well as API keys?
Yes. Scalekit vaults both Prisma API keys and direct Prisma Postgres connection strings. Each is stored and resolved the same way, giving you consistent per-user isolation and audit coverage whether your agent connects through the Prisma Data Platform or a direct database URL.
.svg)
.svg)
