Update project name, domain whitelist, or attribution settings. Use only after the user approves changing project configuration.
| Name | Required | Description | Default |
|---|---|---|---|
| name | No | ||
| projectId | No | Optional project ID. Omit only when the API key is project-scoped or the account has a clear default project. | |
| allowedDomains | No | Optional full replacement list of allowed hostnames for the project. | |
| attributionModel | No | ||
| attributionWindowDays | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations indicate it's a mutation (readOnlyHint false) but not destructive. The description adds the important context that user approval is required. However, it doesn't disclose what happens to omitted fields (e.g., do they remain unchanged or reset?) or whether the operation has side effects, leaving gaps beyond 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 extremely concise—just two sentences that front-load the action. Every word adds value with no redundancy or 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?
With no output schema, the description should cover return values or success indicators, but it doesn't. It also omits how partial updates work (e.g., whether providing only name leaves other fields unchanged). More context is needed for safe usage.
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 low (40%), so the description should compensate. It does group fields into categories but provides no additional meaning for attributionWindowDays or the behavior of projectId beyond the schema. This is insufficient given that 60% of parameters lack descriptions.
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 it updates specific fields: name, domain whitelist, and attribution settings. It effectively distinguishes from sibling tools like create or archive by focusing on existing project modifications, though it doesn't explicitly name alternatives.
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 a key condition: use only after user approval. However, it lacks guidance on when not to use it (e.g., creating a new project) and does not mention alternative tools like projects_create or projects_update in different contexts.
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.
Each tool targets a distinct resource and action combination (e.g., domains_add, experiments_create, goals_deactivate, reports_experiment_chart). There is no overlap or ambiguity; even the three report tools serve clearly different purposes (totals, time-series, channel breakdown).
The vast majority of tools follow a resource_action snake_case pattern (domains_add, experiments_list). A couple deviate (billing_portal, usage_summary) but still place the resource first, making the pattern predictable and easy to parse.
With 23 tools covering projects, domains, experiments, goals, reports, billing, health, and usage, the count is well-scoped for a split-testing platform. Each tool addresses a specific need without ballooning into excessive granularity.
The tool surface provides CRUD-like operations for core entities (projects, experiments, goals, domains) and essential report types. Minor gaps exist (no goal update tool, no experiment delete—only archive) but these are reasonable trade-offs for the domain.
