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

Beyond the annotations (destructiveHint, openWorldHint, non-idempotent), the description discloses that writes may return a link the user must approve, that repeating the same publish or comment within 15 minutes returns the first result instead of double-writing, that Instagram throttles bursts, and that on GMB delete removes only the owner reply. It also notes validate_only and that results carry a next_action to follow — all behavior an agent could not infer from the schema.

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?

Purpose and platform codes are front-loaded, followed by action-by-action parameter mapping, so the structure is logical and every clause carries information for a 37-parameter tool. It is still a dense semicolon-chained wall of text that is harder to scan than it needs to be, which is the only thing holding it below a 5.

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 seven actions, three platforms, 37 parameters and a mutation-heavy annotation profile, the description covers the decision points that matter: platform selection, id formats, approval flow, dedup window, throttling, and the follow-next_action convention. Return values are covered by the output schema, so the description need not explain them.

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 89%, so the baseline is 3, but the description goes further by mapping parameters to actions (all take ds_id; media_id vs comment_id to reply; hidden/delete only for moderate_comment) and by explaining LinkedIn id formats (share-<n>, ugcPost-<n>, post URN, update_id) and where GMB account_id and comment ids come from. It stops short of covering every field (e.g. job_id, wait_seconds are only implied), which keeps it from a 5.

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 a concrete domain (comments, publishing, scheduling) and enumerates the seven actions it dispatches (list_comments, comment, moderate_comment, publish, update_post, cancel_scheduled, status) across three named platforms with their ds_id codes. It explicitly routes adjacent jobs elsewhere (creative_picker for AI images, accounts_discovery for GMB location ids), so an agent can separate it from siblings without opening the schema.

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?

It states when to use each path rather than leaving it to inference: generate AI images in creative_picker first, use media_prompt only if the user declined review, omit ds_id for scheduled posts/jobs/approvals, and use campaign_and_resource_get for existing saved assets. It also gives a positive precondition (include today, e.g. last_7_days_inc, for new posts) and a scheduling constraint (space one account's posts 5+ minutes apart).

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