Skip to main content
Glama

Connect a source

add_source
Destructive

Connect a SQLite, CSV, Excel, PostgreSQL, or MySQL source with a name for read-only querying; user-entered passwords go to the keychain, never the chat.

Instructions

Connect a source; adding a name again replaces it. The file kinds take path, the file's location, and are added only once the user says yes themselves (Portward asks them): 'sqlite' a SQLite file, 'csv' a CSV file (one table named after the file), 'excel' an .xlsx workbook (one table per sheet). 'postgres' and 'mysql' take host, port, database and user, never a password: the user types the password themselves where you can't see it, and it goes straight to the computer's keychain. Never ask for a password or connection string in the chat.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "name": "<name>",
      "kind": "<kind>"
    },
    "name": "add_source"
  }
}

Input Schema

NameRequiredDescriptionDefault
hostNo
kindYes
nameYes
pathNo
portNo
userNo
databaseNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A3.8/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The annotations flag destructiveHint=true and readOnlyHint=false, and the description adds the crucial detail that explains why: re-adding an existing name replaces it. It also discloses how secrets are handled — the user types the password where the agent cannot see it and it goes directly to the keychain — which is behavioral information the annotations cannot convey. It stops short of describing what happens on failure or what a successful connect returns, keeping it at a 4 rather than a 5.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The definition front-loads the action and the replacement caveat, then organizes the rest by kind, which is the right structure for a tool with mode-dependent parameters. The prose is dense and slightly run-on with stacked quotes and parentheticals, but essentially every clause conveys operational information rather than filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a seven-parameter, destructive, no-output-schema tool with zero schema descriptions, the definition covers the essentials an agent needs: valid kinds, which parameters belong to which kind, consent requirement, replacement semantics, and secret handling. What is absent is the return value or success signal after connecting, and any guidance on error or duplicate handling, which leaves a small but real gap.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With 0% schema description coverage, the description carries the full burden and mostly does: it assigns 'path' to the file kinds, assigns host/port/database/user to 'postgres' and 'mysql', implies the valid 'kind' set, and defines 'name' through the replacement rule. Minor gaps remain — no format or default hints for 'port' (whose schema oddly permits a string), and no distinction between 'database' for postgres vs mysql — so it is strong but not exhaustive.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a concrete action and resource ('Connect a source') and goes further by enumerating the accepted 'kind' values ('sqlite', 'csv', 'excel', 'postgres', 'mysql') with what each one means. It does not explicitly contrast itself with siblings like test_source, describe_source, or list_sources, but the connecting/adding framing is unambiguous enough to distinguish it from read or removal tools.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

There is meaningful usage context: file sources are added only after the user confirms ('added only once the user says yes themselves'), and the agent is told never to ask for a password or connection string in chat. What is missing is routing guidance against the named siblings — nothing says to call test_source after adding or describe_source to inspect an existing one, so the agent must infer the surrounding workflow.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.