connect_todoistConnect Todoist. For security your API token is entered directly in Local MCP, never passed through the AI: LMCP's icon → My apps → Connect, in Todoist's row (with no row: LMCP's icon → My apps → Add apps…). Call this to get these steps, or to check whether Todoist is already connected.
| Name | Required | Description | Default |
|---|---|---|---|
| confirm | No | Ignored; kept for older clients. |
| Name | Required | Description | Default |
|---|---|---|---|
| ok | Yes | ||
| message | Yes |
Changes observed during successful MCP inspections.
Input schema / properties / confirm / descriptionPrevious value: -"Must be true to check the connection and get the steps. Without it the call returns a preview and does nothing."New value: +"Ignored; kept for older clients."Input schema / properties / confirmAdded value: +{
+ "description": "Must be true to check the connection and get the steps. Without it the call returns a preview and does nothing.",
+ "type": "boolean"
+}Input schema / properties / api_tokenRemoved value: -{
- "description": "Your Todoist API token (a 40-char hex string)",
- "type": "string"
-}Input schema / requiredPrevious value: -[
- "api_token"
-]New value: +[]Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description frames this as an informational call — retrieve steps or check status — and explicitly says the token is entered by the user in LMCP, "never passed through the AI." The annotations declare readOnlyHint=false and destructiveHint=true, which directly conflicts with that read-only framing, leaving an agent unable to know whether the call mutates anything.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The exact menu path is genuinely useful and the sentences are dense with actionable detail rather than filler. The second sentence is long and slightly run-on, but no clause is wasted.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
An output schema exists, so return values needn't be described, and the setup path is fully specified. However, the description never resolves the mutation question raised by the annotations, nor what the manual path means for a user who cannot access the LMCP UI — gaps that matter for correct invocation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% and the single parameter is self-documenting ("Ignored; kept for older clients"), so the schema carries the load. The description adds no information about the parameter, which is acceptable at this coverage level but nothing more.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The first sentence says only "Connect Todoist," while the body reveals the tool actually returns the manual connection steps or reports whether Todoist is already connected. The verb/resource is stated, but the real behavior is muddied by a name that implies the tool performs the connection. It does not clearly distinguish itself from siblings like disconnect_todoist or todoist_list_projects.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
"Call this to get these steps, or to check whether Todoist is already connected" gives an explicit trigger for invoking the tool. It does not name alternatives or state when not to call it (e.g. once connected, or when the user wants to disconnect), which keeps it below a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.