Record durable knowledge

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

With no annotations, the description carries the full behavioral burden and falls well short: it never states this is a write/mutation, whether records are deduplicated, what confirmation is returned, or any permission requirements. The account-wide hint it does offer is already stated in the schema's `project` description, so it adds no new behavioral disclosure.

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?

Two tight sentences, no filler: the first defines the qualifying knowledge domain and the second front-loads the account-wide scoping rule. Every clause earns its place.

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?

A seven-parameter mutation tool with no annotations and no output schema gets only ~20 words. An agent cannot tell from the description whether adding is idempotent, what happens to referenced observations, or how insertion relates to update_context/delete_context — substantial gaps for a write operation.

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 57%, so the description must pay some of the freight; it does explain the omit-project/cwd scoping semantics, but that duplicates the `project` parameter's own schema description. The other four parameters (body, title, model, from_observation_id) get no clarification, so the description neither compensates for the coverage gap nor adds much beyond 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 names the resource precisely — durable knowledge of four kinds (decision, convention, gotcha, preference) that map to the `kind` enum, so an agent knows what qualifies for insertion. However, it never states a verb explicitly and does not distinguish itself from siblings like update_context or get_context, so selection still relies on the tool name.

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 only guidance is a scoping shortcut ('omit `project` and `cwd` to make it account-wide'), which is invocation detail rather than when-to-use guidance. Nothing tells the agent when to record knowledge at all, or whether this is preferred over update_context, which is a real ambiguity given the sibling list.

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