agentleFS
nainishshafinainishshafi/developer-productivity-skills/.github/skills/sql-connector/SKILL.md

sql-connector

Skill1 starsChanged 4 days agoMITView on GitHub

Connects to, introspects, and queries multi-engine databases (Microsoft SQL Server, Oracle, PostgreSQL, SQLite, MongoDB, MS Access, MySQL) via the sql-connector TypeScript MCP server. Enforces zero-tolerance safety guardrails preventing DROP TABLE, DELETE * or unbounded updates, and destructive DDL. Use when asked to "query database", "connect to sql server", "query postgres", "query oracle", "inspect database schema", "connect to sqlite", "query mongodb", "connect to ms access", "run safe sql", or "audit sql query safety".

  • Deletes or force-pushes

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every query is filtered through the SQL Connector Firewall before execution:

    line 29

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

SQL Connector

Safely inspect database schemas, evaluate query plans, and execute guarded SQL or MongoDB queries across relational and document databases using the sql-connector MCP server.


Architectural Heuristics & Decision Matrix

Database EngineRecommended AdapterSafety Mode GuidancePrimary Use Cases
SQLiteBuilt-in node:sqlite (Node 22+)read-only (default) or guarded-writeLocal testbeds, isolated reproduction fixtures, scratch databases.
PostgreSQLPure-JS pg adapterread-only for audits; guarded-write for bounded data updatesCloud databases, analytical queries, JSONB document querying.
SQL ServerPure-JS tedious (TDS protocol)read-only with maxRows cappingEnterprise transactional databases, T-SQL queries, stored procedure calls.
Oracleoracledb Thin Clientread-only for inspection; EXPLAIN PLAN for tuningEnterprise ERP/core systems, PL/SQL queries, data dictionary audits.
MongoDBmongodb driver (MQL)Structured JSON queries (find, aggregate)Document stores, nested JSON collections, event streams.
MS AccessWindows ACE OLEDB / ODBCread-only schema and table discoveryLegacy desktop databases (.accdb, .mdb), local file data extraction.
MySQL / MariaDBPure-JS mysql2read-only (default)Web applications, read replicas, relational queries.

Zero-Tolerance Safety Guardrails

Every query is filtered through the SQL Connector Firewall before execution:

  1. Forbidden DDL: DROP TABLE, DROP DATABASE, TRUNCATE TABLE, ALTER TABLE ... DROP are strictly blocked in all modes.
  2. Forbidden DML: DELETE *, DELETE FROM table without WHERE, and tautological clauses (WHERE 1=1, WHERE 'a'='a', WHERE true) are blocked.
  3. Execution Mode: Defaults to read-only (only SELECT, EXPLAIN, SHOW, find). Use guarded-write strictly for bounded INSERT/UPDATE/DELETE with explicit targets.
  4. Context Protection: Results are capped to maxRows: 500 by default to prevent context window exhaustion.

Modular References

  • SQL Connector MCP Reference Guide — Tool specifications, connection strings, firewall rules, and dialect examples.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.