Build the connection in stages: let Antigravity call SQLcl’s MCP tools, let SQLcl connect through a saved database profile, and give that profile only the permissions the workflow needs. Start with a read-only query. Add Oracle AI Agent Memory only if you need context to persist across sessions; it complements live database access rather than replacing it.
How the workflow fits together
Antigravity does not connect directly to Oracle Database in this setup. It acts as the MCP client: it discovers and calls tools exposed by SQLcl, and SQLcl executes operations through a named or saved connection. The database remains the source of data and, when configured, the durable store for memory and retrieval evidence.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
Oracle SQL and Pl/Sql | $50.50 | Buy on Amazon |
| 2 |
|
Oracle PL / SQL For Dummies | $15.95 | Buy on Amazon |
| 3 |
|
Mastering Oracle SQL, 2nd Edition | $20.80 | Buy on Amazon |
| 4 |
|
Oracle PL/SQL by Example (The Oracle Press Database and Data Science) | $48.81 | Buy on Amazon |
| 5 |
|
Oracle PL/SQL Programming: Covers Versions Through Oracle Database 12c | $63.84 | Buy on Amazon |
As an Amazon Associate I earn from qualifying purchases.
- Antigravity: the developer-facing agent interface and MCP client.
- SQLcl MCP: the tool boundary between Antigravity and SQLcl connections and operations.
- Oracle AI Database: the durable database layer for application data, retrieval evidence, vectors, metadata, and traces.
- Oracle AI Agent Memory: optional application APIs for threads, durable memories, scoped retrieval, and context assembly.
- LangChain: an optional application-side layer for reusable retriever, document, or chain orchestration.
There are two distinct loops. In the live interaction loop, Antigravity calls SQLcl MCP tools and receives results from SQLcl’s database operations. In a separate memory loop, application code can store histories, tool logs, memory records, chunks, and embeddings in Oracle AI Database, then retrieve scoped context for later steps. SQLcl MCP supplies live tool access; memory and retrieval support continuity and context assembly.
Free tools Windows power users keep installed
One-click scans. No signup required.
Oracle’s guide describes a companion notebook as a build-and-validation harness. It checks SQLcl and Java discovery, previews sanitized MCP configuration, validates a saved connection alias, creates memory tables, inserts simulated traces, exercises lexical, vector, and hybrid retrieval, initializes the memory package, and captures a validation snapshot. Treat it as a reference workflow, not as evidence of independent testing or performance results.
#1 Best Overall
Minimum setup and prerequisites
Oracle’s September 18, 2026 guide specifies SQLcl 25.2.0 or later, JRE 17 or 21, Antigravity MCP configuration, a saved SQLcl profile with password persistence, and a database user limited to the workflow’s needs. Oracle’s SQLcl documentation is now at version 26.1; configuration keys, tool names, and arguments can vary by release, so check the instructions for the versions you actually install. See the SQLcl 26.1 documentation and Oracle’s Antigravity workflow guide.
- Install SQLcl and an appropriate JRE; verify SQLcl with
sql -Vand confirm Java is discoverable. - Use an approved development, replica, or other sanitized database environment for initial validation.
- Create a dedicated database account with only the permissions needed; begin read-only.
- Decide how the local saved-password store will be protected under your organization’s secrets policy.
- Configure Antigravity to launch the installed SQLcl executable as an MCP server.
Build and validate the local SQLcl MCP connection
- Install and check the runtime. Install SQLcl 25.2.0 or newer and JRE 17 or 21, then run
sql -V. Confirm the Java runtime is available to the SQLcl process. - Create a restricted database account. Grant only the read access needed for the first query. Avoid using an account with broad administrative privileges to validate the integration.
- Save and test a named SQLcl connection. Oracle’s example uses
conn -save antigravity_mcp -savepwd <ORACLE_USER>/<ORACLE_PASSWORD>@<ORACLE_DSN>. Substitute your approved credentials and database descriptor, and test the alias directly in SQLcl before configuring Antigravity. Do not ask the agent to invent credentials at runtime. - Add SQLcl as an MCP server in Antigravity. Configure the server entry to launch the absolute path to the SQLcl executable with
-mcp. Oracle’s example usesmcp_config.json; use the configuration format and location supported by your installed Antigravity version. Keep passwords out of the configuration and any preview or shared file. - Reload and verify tool discovery. Reload the MCP client and check that the SQLcl tools appear. Then use Antigravity for one bounded, read-only query against the saved alias, with a result limit appropriate to your test.
- Review activity before increasing access. Inspect SQLcl and database activity and logging. Record useful audit context—tool, identity, timestamp, status, and sanitized input/output—without exposing secrets or unnecessary sensitive data.
The exact Antigravity configuration keys and SQLcl tool names are release-specific; the Oracle article provides a versioned example rather than a guarantee that every installed release uses identical syntax. Consult current Oracle guidance and your Antigravity release documentation before copying configuration verbatim.
Rank #2
What MCP does—and what it does not secure
MCP makes the available tool boundary explicit, but it is not an authorization system. The effective access comes from the saved SQLcl connection, database user, grants and roles, network controls, and database policies. A tool call cannot be made safe merely by prompting the agent not to perform an operation if the underlying account can perform it.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
- Use a dedicated, least-privilege account and start with read-only grants.
- Keep development, test, and production identities and policies separate.
- Protect stored credentials according to local secrets-management rules.
- Review database activity and tool logs, and require explicit approval for risky operations before expanding permissions.
Oracle’s workflow guide emphasizes boundaries, privileges, logging, scoped retrieval, and repeatable runbooks as production concerns. Those controls belong around the database connection and application, not in MCP alone.
Rank #3
When to add Oracle AI Agent Memory or LangChain
Oracle AI Agent Memory for durable, scoped context
Add Oracle AI Agent Memory when an application needs thread context to persist between sessions, durable memories, scoped recall, or assembled context for later answers. In this pattern, Oracle AI Database stores the memory and retrieval data. It is an additional application capability, not a prerequisite for Antigravity to call SQLcl MCP.
LangChain for application-side orchestration
LangChain is optional. Consider it when your application already needs reusable retriever, document, or chain orchestration. It is not required simply to connect Antigravity to SQLcl MCP.
Embedding and LLM providers
The notebook’s default path uses a local deterministic embedder, according to Oracle’s guide. A provider key is needed only if you modify that path to call provider-backed embedding or LLM services. That distinction applies to the notebook workflow described by Oracle, not necessarily to every application built on these components.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteChoose a deployment model
Local SQLcl MCP is one option, not the only way to expose Oracle database tools. Oracle describes three patterns; confirm current availability and environment-specific details before selecting one.
| Option | Deployment and identity | Best fit described by Oracle |
|---|---|---|
| SQLcl MCP | Local SQLcl process using saved SQLcl connections. | Local development, prototyping, and individual developer productivity. |
| OCI Database Tools MCP | Managed, serverless OCI service with OCI IAM integration. | Centrally managed access to Oracle cloud databases. |
| ORDS MCP | ORDS Standalone with an HTTPS streaming /mcp endpoint, database connection pools, and OAuth-related identity integration. |
Teams with existing ORDS deployment patterns and a chosen identity provider. |
The practical choice depends on where the MCP server runs, who administers it, and how client and database identities are authenticated. Oracle’s MCP overview describes the deployment approaches; check its current documentation for feature and availability details for your environment.
Quick Recap
A sensible rollout after the first query
- Keep the first working setup read-only and bounded; review the returned data and corresponding database activity.
- Document the approved connection alias, account grants, network path, logging expectations, and recovery owner in a repeatable runbook.
- Introduce durable memory only when a concrete cross-session or scoped-recall requirement exists, and define what may be stored and retrieved.
- Expand permissions or move to a managed MCP deployment only after reviewing identity, administration, audit, and operational requirements.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
