startStart a compute instance / resource identified by its id - Starting a compute instance / resource is like powering on a real server. If the compute instance / resource is already started nothing will happen. You may check the current status anytime when getting information about a compute instance / resource.
| Name | Required | Description | Default |
|---|---|---|---|
| instanceId | Yes | ||
| x-trace-id | No | ||
| x-request-id | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already indicate a mutating, potentially destructive action (readOnlyHint=false, destructiveHint=true). The description adds valuable context: the idempotent behavior ('nothing will happen' if already started) and the analogy to powering on a server. It does not contradict the 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?
The description is front-loaded with the core action, but it repeats 'compute instance / resource' multiple times unnecessarily. The dash and two additional sentences are acceptable, but the writing is verbose and could be tightened without losing meaning.
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 simple start operation with annotations and a schema, the description covers the main behavior and idempotency. However, it lacks any mention of required headers, error handling, or asynchronous behavior, leaving gaps in the agent's understanding of how to invoke it successfully.
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 0%, so the description must compensate for parameter documentation. It only vaguely refers to 'its id', not clearly mapping to the instanceId parameter, and entirely omits the required x-request-id and optional x-trace-id headers. The schema provides types but no semantic explanation.
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 the action ('Start') and the resource ('compute instance / resource identified by its id'). It distinguishes from sibling tools like 'restart', 'stop', and 'rescue' by conveying the power-on semantics.
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?
The description provides clear usage context: powering on a server and idempotent behavior (if already started, nothing happens). It also suggests checking status via the retrieve operation. However, it does not explicitly mention when not to use it or point to alternatives like 'restart' for rebooting.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Many tools have clearly distinct CRUD roles per resource, but there are several confusing overlaps: start/stop/restart/shutdown/rescue are similar lifecycle actions, and there are duplicate audit tools like retrieveImageAuditsList and retrieveImageAuditsList1 (one even mislabeled as DNS Zones audit). Additionally, retrieveTagAuditsList appears to duplicate retrieveAssignmentsAuditsList, and updateDnsZoneRecord incorrectly says 'Create resource record'. These overlaps and mislabeled descriptions make it hard to pick the right tool.
The naming is a mix of camelCase verb_noun patterns, but inconsistent. Some tools use 'retrieveXList' while others use 'listX', some use 'Cancel' with a capital letter, and 'tool_search' uses snake_case. Numeric suffixes like 'retrieveImageAuditsList1' and verbs like 'patchInstance' instead of 'updateInstance' further break consistency.
With 124 tools, this is an extremely large tool surface for an MCP server. Even for a broad cloud provider API, this overwhelms an agent's context and selection capabilities. The number far exceeds reasonable scoping and creates unnecessary selection overhead.
The tool set covers a wide range of resources thoroughly: instances, snapshots, images, private networks, object storage, DNS, domains, tags, roles, users, secrets, and audits. Most resources have create/retrieve/update/delete lifecycle operations, and there are extensive audit history tools. Minor gaps exist (e.g., no explicit instance deletion besides cancel, no separate firewall management beyond upgradeInstance), but overall the surface is quite complete for its domain.