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

Annotations already flag destructiveHint=true and idempotentHint=true, but the description adds real context beyond them: each change is retained in the brain's history, `force` overrides a fact that changed since the proposal, and brain:write plus the Autopilot plan are required. It stops short of describing failure modes when items are stale or conflicting, which matters for a destructive mutation.

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 most important information — what the call means (the founder's explicit yes) and that history is preserved — is front-loaded, and every sentence carries substantive content. The inline backtick-parameter enumeration makes the single paragraph dense, but there's no filler or repetition.

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?

For a destructive, approval-gated write tool with a rich nested schema and no output schema, the description covers meaning, permissions, plan requirement, history retention, and the return value ('the revision as it now stands'). The gaps are the undocumented `id`/`startupId` parameters and the absence of guidance on partial failures or conflicting items.

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 description coverage is 0% across 6 parameters, so the description carries the burden and does most of it well — it defines `itemIds` including the 'all' semantics ('every proposed item but the unclear ones'), `edits` as the founder's own value or text, `rejectItemIds`, and `force`. The two remaining parameters, `id` and `startupId`, are never mentioned, and the mapping of edit payload shapes (value vs text vs scope) to fact/answer/guideline types is only implied.

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 opens with a specific verb+resource: applies items of a proposed revision to the company brain, and lists what those items can be (accepted, edited, rejected, forced). It's clear what the tool does, but it never names the sibling that creates the proposal (getfund_brain_revise) or the one that reads it (getfund_brain_revision), so the agent must infer the workflow split from context.

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?

Usage context is stated clearly: the call represents the founder's explicit yes, given only after they saw each item's current and proposed value, and the required plan and scope are named. What's missing is explicit routing guidance such as 'to propose changes, use X; to inspect the proposal first, use Y' and any statement of when not to call (e.g. no pending revision).

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