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

Annotations already cover readOnly/idempotent/non-destructive/openWorld, so the bar is lower, yet the description adds real behavior: results are grant-filtered ('nothing you cannot read is listed or counted'), 'people: ... names only, never content', and spans/truths flag staleness 'because a truth they cite was superseded'. These are non-obvious traits not derivable from the schema or 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?

Only the first clause is front-loaded; the remainder is a single dense run-on of em-dash-separated action clauses that is hard to scan. Given 20 actions each sentence arguably earns its place, but the structure could be far more navigable (e.g., one line per action).

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 20-action, 16-parameter dispatcher with an output schema present, the description maps arguments to actions and flags scope/permission behavior, so an agent has enough to choose an action and its inputs. It does not, however, address paging across actions generally (only documents mentions limit/offset) or return shape per action beyond what the output schema provides.

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, but the description adds semantics the schema lacks: lessons query must be 'a key term rather than a question' and results are 'ranked by how many independent chains found it useful', and label matching is described as including inherited labels. These go beyond restating the parameter docs.

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?

Opens with a specific verb+resource+scope: 'List what you can see, filtered to your grants,' and then enumerates each of the 20 actions with its own subject matter. It also distinguishes itself from siblings by routing reads to 'read action=document'. The breadth of a 20-action dispatcher keeps it from being crisply singular, but an agent can tell what this tool returns per action.

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?

Offers explicit disambiguation for the trickiest overlap: 'sources: the CONNECTIONS to GitHub and Google Drive, NOT documents in a folder (action=documents lists those)'. It also tells the agent to use a sibling ('read one with read action=document') for retrieval. It stops short of stating when browse is preferable to the sibling 'search' or 'read' for listing, so it is not a full when/when-not guide.

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