Interfaces with databases to run SQL queries, view tables, and inspect schemas.
Install
mkdir -p .claude/skills/whodb && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/6569" && unzip -o skill.zip -d .claude/skills/whodb && rm skill.zipInstalls to .claude/skills/whodb
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Database operations including querying, schema exploration, and data analysis. Activates for tasks involving PostgreSQL, MySQL, MariaDB, TiDB, SQLite, MongoDB, Redis, Elasticsearch, or ClickHouse databases.Key capabilities
- →List available database connections
- →Execute SQL queries
- →Explore database schemas and table structures
- →Describe table columns and relationships
- →Export query results to CSV or Excel
How it works
The agent interacts with databases via MCP tools or CLI commands to perform schema discovery and execute targeted SQL queries.
Inputs & outputs
When to use whodb
- →Query SQL databases for data analysis
- →Inspect table schemas and column details
- →List connections and available tables
- →Execute SQL migration checks
About this skill
WhoDB Database Assistant
You have access to WhoDB for database operations. Use these tools and commands to help users with database tasks.
Platform MCP (default)
whodb mcp serve exposes hosted platform tools. Sign in to each required host with
whodb login --host <url>. Start with whodb_platform_hosts to discover saved
hosts and accounts, then whodb_platform_orgs with workspace: {host} and
whodb_platform_projects with workspace: {host, org}.
When the user wants the whole conversation on one host (for example UAT or
production), call whodb_platform_session_use once with
workspace: {host, org, project}. It verifies the login, resolves canonical IDs,
and becomes the default for later calls in this MCP session only. It is never
saved to disk, so other chats and terminals keep their own defaults. Confirm
with whodb_platform_status, which reports session_scope. Use
whodb_platform_session_clear to drop it.
For one-off calls to another host or project, pass an explicit
workspace: {host, org, project} on that call. Resolve it first with
whodb_platform_workspace_resolve if unsure. Per-call targets, including
parallel calls, never change the session default or saved defaults. Check the
returned scope. Resolution order for an omitted workspace: session default,
then process overrides, then saved defaults. Do not run whodb use to switch
hosts for one chat; it rewrites the saved default for every session. Read
whodb://platform/schema for operations and payloads.
To answer a question about ontology data, call whodb_platform_ontology_describe
first to learn entity and property names, then whodb_platform_ontology_aggregate
for exact counts, sums and group-bys, or whodb_platform_ontology_query for
filtered rows, search and joins (bounded scan, max 5000 rows). Drill into rows
with whodb_platform_ontology_rows. For bulk inserts use
whodb_platform_add_ontology_records with an idempotency_key. Preview behavior
actions with whodb_platform_ontology_action_preview before executing them.
Writes return a preview bound to its host, account, organization, and project.
Confirm only after approval of that exact preview. Changing defaults does not
redirect a pending confirmation. Ask for whodb login --host <url> when a host
needs authentication; no separate login is needed per project.
Guided setup
This skill is optional. WhoDB MCP supplies its own instructions, tool schemas,
and setup guidance. Prefer live MCP discovery if this skill differs from it.
Keep one entry: command whodb, args ["mcp", "serve"].
Use whodb_mcp_setup with action: "inspect" to discover settings, targets,
choices, and the settings schema. Ask the user what they want, then use
action: "preview" with a patch. Show the proposed settings and obtain approval
before action: "apply" with the returned confirmation_token and approved: true.
Do not send a patch with apply. Tokens expire after five minutes; only the latest
preview is valid. Restart MCP after changed settings, then use action: "verify".
Verification checks active settings. Omitted patch fields are preserved.
No shell access is required for this flow. Browser sign-in and credential entry
remain user actions; never request credentials in chat. If setup tools are
filtered out, ask the user to run whodb setup. CLI agents can alternatively use
whodb setup inspect, validate, apply --yes, and verify.
Inspect whodb://mcp/configuration for active modules and restrictions.
Platform calls target workspace; database calls target connection.
Database-only MCP
Enable the database module through whodb setup; both modules can run in the same whodb mcp serve connection. --database remains an advanced process override.
This connects directly to databases without a WhoDB platform server; databases
can be remote. The terminal UI is available with whodb --tui.
Database tools
When database-only MCP is configured, use these tools directly:
whodb_connections
List all available database connections.
No parameters required.
Returns: List of connection names with type and source (env/saved).
whodb_query
Execute SQL queries against a database.
Parameters:
- connection: Connection name (optional if only one connection exists)
- query: SQL query to execute
Example: whodb_query(connection="mydb", query="SELECT * FROM users LIMIT 10")
whodb_schemas
List all schemas in a database.
Parameters:
- connection: Connection name (optional if only one connection exists)
Example: whodb_schemas(connection="mydb")
whodb_tables
List all tables in a schema.
Parameters:
- connection: Connection name (optional if only one connection exists)
- schema: Schema name (optional, uses default if not specified)
Example: whodb_tables(connection="mydb", schema="public")
whodb_columns
Describe columns in a table.
Parameters:
- connection: Connection name (optional if only one connection exists)
- table: Table name (required)
- schema: Schema name (optional)
Example: whodb_columns(connection="mydb", table="users")
CLI Commands (Fallback)
If MCP tools are unavailable, use the CLI directly via Bash:
Query Execution
whodb query "SELECT * FROM users LIMIT 10" --connection mydb --format json
Schema Discovery
# List schemas
whodb schemas --connection mydb --format json
# List tables
whodb tables --connection mydb --schema public --format json
# Describe columns
whodb columns --connection mydb --table users --format json
Connection Management
# List connections
whodb connections list --format json
# Test connection
whodb connections test mydb
# Add new connection (interactive)
whodb connections add --name mydb --type Postgres --host localhost --database mydb
Data Export
# Export to CSV
whodb export --connection mydb --table users --output users.csv
# Export query results
whodb export --connection mydb --query "SELECT * FROM orders" --output orders.xlsx
Workflow Examples
Explore a New Database
- List connections:
whodb_connections - List schemas:
whodb_schemas(connection="name") - List tables:
whodb_tables(connection="name", schema="public") - Describe table:
whodb_columns(connection="name", table="users") - Sample data:
whodb_query(connection="name", query="SELECT * FROM users LIMIT 5")
Answer Data Questions
- Understand the schema first - check table structure
- Write targeted queries with appropriate filters
- Always use LIMIT for exploratory queries
- Present results in a clear, readable format
Best Practices
- Always explore schema first before writing queries
- Use LIMIT for exploratory queries to avoid overwhelming output
- Prefer specific columns over SELECT * for clarity
- Check foreign keys via whodb_columns to understand relationships
- Use JSON format (--format json) when parsing output programmatically
- Never expose credentials - use connection names, not connection strings
When not to use it
- →Exposing raw credentials in queries
Prerequisites
Limitations
- →Requires LIMIT on exploratory queries
- →Must use connection names instead of credentials
How it compares
It provides a standardized interface for database exploration and query execution that avoids manual connection string management.
Compared to similar skills
whodb side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| whodb (this skill) | 1 | 7mo | Review | Beginner |
| database | 0 | 5mo | No flags | Advanced |
| design | 0 | 5mo | Review | Intermediate |
| database-schema-design | 0 | 6mo | No flags | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by clidey
View all by clidey →You might also like
database
asnk633
Database development and operations workflow covering SQL, NoSQL, database design, migrations, optimization, and data engineering.
design
Hainrixz
Greenfield database design — pick the right engine for a new project, compare it against the boring default, and hand back a starter data model with a diagram. Recommendation mode (module M0); never scored, never destructive. Use when the user is starting fresh and asks what database to use, how to
database-schema-design
RepairYourTech
Design database schemas with normalization, relationships, and constraints. Use when creating new database schemas, designing tables, or planning data models for any database paradigm.
data-sql
nholder88
>-
sql-queries
anthropics
Write correct, performant SQL across all major data warehouse dialects (Snowflake, BigQuery, Databricks, PostgreSQL, etc.). Use when writing queries, optimizing slow SQL, translating between dialects, or building complex analytical queries with CTEs, window functions, or aggregations.
database-design
davila7
Database design principles and decision-making. Schema design, indexing strategy, ORM selection, serverless databases.