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

Substantially exceeds the annotations (idempotentHint/destructiveHint/openWorldHint): it discloses that no target agent or connection credential is created, that validation happens before allocation, that partial success is retained rather than rolled back, and that no durable journal exists over remote MCP so the saved plan/operationKey plus service receipts are the recovery authority. That is real behavioral context an agent cannot get 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?

Purpose is front-loaded in the first sentence and the remaining sentences each carry a distinct constraint (aliasing, validation/partial success, retry protocol, credential scope, journal behavior). Sentences like the 'service idempotency receipts are the recovery authority' line are jargon-dense, but none are pure filler.

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 complex, deeply nested plan schema with no output schema and only 5 parameters, the description supplies the safety, idempotency, and partial-success semantics an agent needs before invoking. The main remaining gap is the shape of a successful/partial response, though 'partial success is retained' gives a usable hint.

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 description coverage is only 20% (essentially just operationKey), so the description must compensate. It adds meaning for alias ('Duplicate Tools use distinct aliases'), the plan/operationKey pair ('Persist exact plan and operationKey'), and dataset validation, but says nothing about source kinds, data kinds, seed, ttlSeconds, or waitTimeoutMs. Partial compensation only.

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?

States a specific verb+resource: 'Start optional named manual Tool instances from pinned catalog releases or saved private builds', including the two source kinds. It distinguishes itself implicitly from creation tools (create_project, service_create) by being a start/instantiate operation, but never names a sibling explicitly, so an agent must infer the boundary.

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?

Provides operational when-guidance ('Persist exact plan and operationKey before calling; retry only that unchanged intent') and implicitly scopes usage to named manual Tools, but never states when to choose this over firedrill_service_create, firedrill_materialize, or create_project. Usage is implied rather than contrasted with alternatives.

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