More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    db explorer

    1

    Read-only SQL database exploration, schema auditing, and query optimization via MCP.

    Free

    2 usesSecurity scanned
    db explorer

    db explorer

    Example session with this skill installed

    Check the schema health of my local postgres database and see if any foreign keys are missing indexes.

    • Read your context and instructions
    • Compiled the db explorer

    I have performed a schema validation on your PostgreSQL database. I identified several foreign keys without corresponding indexes and one table missing a primary key.

    Schema Health Audit

    Issue TypeTable NameColumn/DetailImpact
    Missing Indexorderscustomer_idSlow joins on customer lookups
    Missing Indexorder_itemsproduct_idPerformance hit during inventory analysis
    Missing PKtemp_logsN/ARisk of duplicate rows and slow updates
    Wide Tableuser_profiles52 columnsHigh I/O overhead for simple reads
    Unindexed FKtransactionsvault_idPotential locking issues on parent deletes

    Suggested Indexing SQL

    -- Fix for orders table
    CREATE INDEX idx_orders_customer_id ON orders(customer_id);
    
    -- Fix for order_items table
    CREATE INDEX idx_order_items_product_id ON order_items(product_id);
    
    -- Fix for transactions table
    CREATE INDEX idx_transactions_vault_id ON transactions(vault_id);
    

    Next steps

    1. Review the suggested CREATE INDEX statements in your staging environment.
    2. Run a SELECT count(*) on the temp_logs table to determine if a migration to add a Primary Key is feasible.
    3. Consider normalizing the user_profiles table to reduce the column count per row.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Connecting an AI agent to a live database usually carries significant risk of data loss or performance degradation. This skill provides a safe, read-only interface to PostgreSQL, MySQL, and SQLite databases using a specialized MCP server. It enables deep schema inspection, query execution with row caps, and performance tuning without write permissions.

    What it does

    • Schema Discovery lists tables, views, and specific column details across any supported SQL dialect.
    • Safe Querying executes SELECT statements inside read-only transactions with hard timeouts and row caps.
    • Performance Profiling retrieves query plans and identifies full table scans to pinpoint bottlenecks.
    • Index Suggestions generates CREATE INDEX statements for unindexed foreign keys and slow join columns.
    • Schema Auditing flags missing primary keys, redundant indexes, and wide table structures.
    • Migration Validation checks local SQL migration scripts for unsafe operations before deployment.

    How it works

    1. Inspection starts by mapping the database structure using schema exploration tools to identify relevant tables.
    2. Analysis runs specific SELECT queries to fetch data samples or calculate aggregates directly in the database.
    3. Optimization uses query explanation tools to analyze execution paths and recommend indexing strategies.
    4. Validation runs health checks on the schema or migration files to ensure best practices are followed.

    Frameworks & tools

    This skill requires the db-explorer MCP server, Python 3, and the uv package manager. It supports PostgreSQL, MySQL, and SQLite via standard connection strings.

    Why this beats prompting it yourself

    Writing SQL and pasting schema definitions manually is slow and prone to context window overflow. This skill automates the feedback loop, allowing the agent to fetch exactly the metadata it needs and test queries in a sandboxed, read-only environment.

    Use cases

    • Performance Tuning finding and fixing slow queries in an existing PostgreSQL instance.
    • Data Exploration generating reports and summaries from unknown database schemas.
    • Schema Health auditing a legacy MySQL database for missing constraints and indexes.
    • Migration Prep drafting and validating DDL scripts for a new feature deployment.

    Known limitations

    The server must be installed and running locally with Python 3 and uv. It supports only read-only SELECT operations and cannot execute DDL or DML commands directly.

    Requires db-explorer MCP server installation. Read-only by design; cannot execute writes or migrations. Row results capped at 100. Dialect support limited to PostgreSQL, MySQL, and SQLite.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    2 uses

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Requires

    GitHub
    Mysql
    Postgresql
    Python
    DB Explorer MCP
    Git

    Permissions

    Terminal / Shell
    Network Access
    Environment Variables

    Allowed Hosts

    api.github.com
    mysql.com
    postgresql.org
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    No, this skill is strictly read-only. It uses specialized MCP tools to execute SELECT statements within sandboxed transactions, ensuring your production data remains safe from accidental deletion or modification.

    It supports PostgreSQL, MySQL, and SQLite. You simply provide the standard connection string, and the skill handles the dialect-specific schema mapping and query optimization automatically.

    This skill requires a local installation of Python 3 and the 'uv' package manager to run the MCP server. Once installed, it connects your AI agent to your database via a secure, local bridge.

    The skill analyzes query execution plans to identify full table scans and bottlenecks. It then provides specific recommendations, such as 'CREATE INDEX' statements, to help you resolve performance issues without manual trial and error.

    Unlike manual prompting, this skill allows the agent to dynamically fetch only the relevant parts of your schema. This prevents 'context window overflow' and ensures the agent always has up-to-date metadata without you having to copy-paste table definitions.

    You receive the complete db-explorer MCP server package, comprehensive configuration documentation, and instructions for integrating it with MCP-compatible AI agents and environments.

    New in Data & Databases