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

Annotations already mark this as destructive/non-read-only, but the description adds substantial behavioral context: it spends PostHog Vision credits, returns either an observation_id or a queued workflow_id, and only works if PostHog is connected. It also explains the write-tier approval behavior with clear instructions. This goes well beyond the annotations and gives the agent actionable risk/cost information.

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 core description is front-loaded and efficient, covering action, cost, return, and trigger in four tight sentences. The write-tier approval paragraph is verbose but contains operational instructions essential for invocation. Minor redundancy exists between 'Run... now' and 'Use when you want one recording scored now,' preventing 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?

For a tool with no output schema, the description usefully explains the return value (observation_id or queued workflow_id) plus the credit cost, connection prerequisite, and approval workflow. It could clarify what a queued workflow_id means or how to resolve it, but an agent still has enough to call the tool correctly. Slight gap in explaining downstream handling of the queued case.

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?

The input schema already covers all three parameters with descriptions (companyId, scanner_id, session_id), so schema coverage is 100%. The description reinforces that this acts on 'one session recording' and a 'scanner,' which loosely maps to session_id and scanner_id, but it adds no new parameter-level detail beyond what the schema provides. Baseline 3 is appropriate.

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 clearly states a specific verb/resource/scope: 'Run a Replay Vision scanner against one session recording now.' It also differentiates from sibling PostHog tools by noting this is an execution tool that returns an observation_id or queued workflow_id, unlike list/get/delete scanner tools. This allows an agent to distinguish it without opening schemas.

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 gives an explicit trigger: 'Use when you want one recording scored now' and a prerequisite: 'Only works if PostHog is connected.' However, it does not explicitly name alternative tools or state when NOT to use this one (e.g., when you need batch scoring or want to inspect existing observations), so it stops short of full when/when-not guidance.

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