delete_data_collectionONLY unassigned + zero rows. Unassigned WITH rows is content preserved from a deleted site — reuse it via assign_to_project instead. Anything attached or holding content is refused (409) — relay the reason to the owner, never retry.
| Name | Required | Description | Default |
|---|---|---|---|
| collectionId | Yes | The collection ID (from list_data_collections) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond the destructiveHint annotation, the description reveals the 409 refusal behavior, the reason to relay to the owner, and the no-retry expectation. It also identifies the risky edge case of deleting preserved content.
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?
Three dense sentences, each earning its place: the precondition, the alternative, and the refusal behavior. The key guardrail is front-loaded.
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 single-parameter destructive tool with no output schema, this description provides everything needed to decide and invoke correctly: preconditions, error handling, and fallback routing. No significant gap remains.
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 100%, so the schema already explains collectionId and its source. The description adds no parameter-specific meaning, which is acceptable given the baseline for full schema coverage.
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 identifies the operation as deleting a data collection while adding the precise invariant (only unassigned and zero rows). It distinguishes this from assign_to_project by naming the alternative for collections that still hold rows.
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 explicitly states when deletion is allowed, when it is refused, and the alternative action for unassigned collections with rows. The guidance 'never retry' also tells the agent not to loop on a 409.
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.
Most tools are clearly separated by resource and action, and the descriptions are detailed enough to avoid most misselection. The main ambiguity is between deploy_files and deploy_website (both deploy a site, one multi-file and one single-file), and get_project versus get_project_files versus list_projects require careful reading.
Every tool follows a consistent snake_case verb_noun pattern: create_*, get_*, list_*, delete_*, update_*, upload_*, assign_to_project, unassign_from_project, configure_chatbot. There are no camelCase names, vague verbs, or mixed conventions.
At 29 tools, this exceeds the comfortable MCP surface and crosses into the 'too many' range. The breadth is somewhat justified by the number of subdomains (projects, files, forms, data, bookings, chatbot, media), but several getters and listers could be consolidated without losing capability.
The tool set covers a full website build lifecycle: account checks, content modeling, media upload, deploy, assignment, chatbot configuration, and post-deploy file edits. Minor gaps remain: booking pages and form settings are intentionally read-only via the MCP, and there is no chatbot deletion or teardown tool.