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

Annotations only state readOnlyHint=false and destructiveHint=false, which are minimal. The description adds critical behavioral context: 'Recording the same insight twice replaces your earlier note; a pinned note is left as it is,' revealing the overwrite behavior. It also mentions the required 'act' scope and describes the exact return value ('the top of their board with move ids'), going far beyond the annotations.

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 long but every sentence carries operational weight. It is front-loaded with purpose, then structured with 'WHEN:' and 'WHAT COMES BACK:' markers, making it scannable. A slight deduction for density—some sentences could be tightened without losing value—but overall the structure is strong.

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 write tool with no output schema, the description covers all critical context: the trigger condition, exclusions, persistence semantics (replace, pinned note), required scope, character limits, and return value. The task is complex enough that this thorough disclosure is appropriate and fully complete.

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%, so the baseline is 3. The description adds qualitative guidance beyond the schema: insight should be 'one plain sentence in the founder's own terms, 10 to 500 characters,' which reinforces and shapes the parameter semantics. It also enumerates the four kinds in the 'kind' parameter via examples in the 'WHEN' section, adding context not fully captured in the schema.

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 'Write one durable thing you have learned about this founder or their product into AfterLaunch's Memory,' using a specific verb ('write') and resource ('Memory'). It clearly distinguishes from siblings like get_snapshot and get_kb_page by explicitly saying not to record restatements of those tools.

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?

Provides explicit 'WHEN' guidance: 'the moment the founder tells you something that will still be true next month and that AfterLaunch could not have measured itself.' It also gives exclusions: 'Never record chatter, a passing mood, a restatement of get_snapshot or get_kb_page, or anything you inferred rather than heard.' This is a model of when-to-use and when-not-to-use.

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