firefox_upload

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

Annotations already declare destructiveHint=true, openWorldHint=true, and non-idempotent, so the mutation risk is known. The description adds genuinely new context: the file choice is performed by the user in the extension panel and agent-supplied local paths are ignored, which is important behavioral information. It omits whether the call waits for the user, whether it can time out, and what happens if the user cancels.

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 short sentences, front-loaded with the action, followed by the key constraint. Nothing is padded, though the phrasing 'Ask the user to choose files in the extension panel' is slightly roundabout compared to 'prompt the user to pick files'.

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?

With no output schema, the description carries responsibility for return behavior and it says nothing about whether an upload status/identifier comes back or how to poll firefox_upload_status. For a destructive, open-world, user-gated operation the definition is minimal but not misleading.

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 50%: ref and snapshotId are documented in the schema, while tabId and frameId are bare integers with no description anywhere. The description does not compensate — 'observed file input' only loosely implies ref/snapshotId and gives no meaning to tabId, frameId, or their interaction (e.g., which frame's input is targeted).

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?

Specific verb + resource: 'Ask the user to choose files ... for an observed file input', which clearly identifies a human-in-the-loop file attachment action tied to a page element. It does not, however, name its closest sibling (firefox_upload_status), so the agent must infer the relationship.

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?

'For an observed file input' implies the precondition that a file input element was already found via a page read, and 'Does not read agent-supplied local paths' is a useful exclusion. But there is no guidance on what to do next (check firefox_upload_status), no mention of when not to use it, and no statement of whether this blocks or returns immediately.

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