Did the founder finish connecting the growth data source you handed them a link for? Pass the poll_token start_connection returned. Reports pending (wait poll_after_seconds and ask again), connected, expired (mint a fresh link) or not_found. Idempotent: nothing is spent or consumed. Read-only, free.
| Name | Required | Description | Default |
|---|---|---|---|
| poll_token | Yes | The poll_token start_connection returned. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
While annotations already indicate readOnlyHint=true and destructiveHint=false, the description adds valuable behavioral context: it is idempotent ('nothing is spent or consumed'), free, and reports specific state values. It also implies retry behavior with 'wait poll_after_seconds and ask again', going beyond the annotation baseline.
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 description is three concise sentences, front-loaded with the core question and purpose. Every sentence adds functional information (usage instruction, possible states, idempotency) with no filler or redundancy.
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?
Given the tool's simplicity (one parameter, no output schema), the description fully covers what the agent needs: what it does, how to invoke it, all possible return states, and safety/behavioral notes. There are no significant gaps for selecting and using this tool correctly.
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?
The schema already provides a complete description for the single parameter (poll_token: 'The poll_token start_connection returned.'). The tool description merely repeats this instruction without adding new meaning, so it meets the baseline for high schema coverage without additional value.
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 description uses a specific verb 'poll' with a clear resource 'connection' and states the exact purpose: checking whether the founder finished connecting a growth data source. It lists the possible outcomes (pending, connected, expired, not_found), which clearly distinguishes it from siblings like start_connection or list_connections.
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?
The description gives explicit context: use the poll_token returned by start_connection, and provides guidance for handling the 'pending' state (wait poll_after_seconds and ask again) and 'expired' state (mint a fresh link). It does not explicitly name alternatives, but the usage scenario is so specific that the intended workflow is clear.
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.
Most tools follow a clear action+resource pattern (get_, list_, set_, ship_, etc.) with distinct purposes. A few boundaries could cause confusion, such as get_move vs get_output (both return drafts) and ship_move vs set_output_status (both mark things shipped), though descriptions mitigate this sufficiently.
The overwhelming majority use consistent snake_case verb_noun naming (list_feed, ship_move, get_visibility). The only outlier is 'whoami', a common convention but not verb_noun; otherwise the pattern is highly predictable.
At 41 tools, the surface is well over the 25-tool threshold for 'too many'. While the domain is broad, this volume places significant cognitive load on agents and suggests some consolidation or modularization could help.
Core workflows for moves and outputs are covered (create/list/get/update/ship/skip/undo). However, the tool descriptions reference 'approve_output' as the publishing action, yet that tool is absent from the list, leaving a dead end for channel distribution. There are also no tools for editing competitors or managing Memory pages beyond recording insights.