neuron_webagent_killThe kill switch — flags a run to stop. It's cooperative: the browser extension stops on its next heartbeat (usually within ~12s) and finalizes the run.
| Name | Required | Description | Default |
|---|---|---|---|
| runId | Yes | The agent-run id to kill. |
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already mark the tool as destructive, non-read-only, and idempotent. The description adds valuable behavior beyond those flags: the stop is cooperative, not immediate, and occurs on the browser extension's next heartbeat, usually within ~12s, after which the run is finalized. This is exactly the kind of behavioral context that helps an agent set expectations correctly.
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?
Two compact sentences front-load the purpose and then explain the key behavioral caveat. There is no filler, and every sentence earns its place.
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 one-parameter destructive tool with strong annotations, the description covers the action, target, cooperative timing, and finalization behavior. It does not describe the return value or error behavior when the run is already stopping, but those are secondary given the simplicity and the lack of an output schema.
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 schema already fully documents the only parameter, runId, with 'The agent-run id to kill.' The description reinforces that the target is a web-agent run and that the action is a kill, but it does not need to add more because schema coverage is 100%.
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 tool's purpose: flags a web-agent run to stop, using the 'kill switch' metaphor and the concrete mechanism of stopping on the next heartbeat. It is specific about the verb and resource, though it does not explicitly contrast with the similarly named neuron_webagent_stop sibling.
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 gives useful context: use this to stop a web-agent run, with the caveat that it is cooperative and typically takes about 12 seconds. However, it does not explicitly state when to prefer this over neuron_webagent_stop or mention any exclusions, so alternative selection is left implicit.
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 type, but there is meaningful overlap in messaging entry points (send_message, send_whatsapp, compose_message, bot_api_send) and contact ingestion/sync tools (import_contacts, populate_contacts, sync_whatsapp_contacts). The descriptions help disambiguate, but with 309 tools an agent will frequently need to read closely to pick the right one.
The overwhelming majority of tools follow a consistent verb_noun snake_case pattern: create_*, get_*, list_*, update_*, delete_*. Minor deviations like sales_stats, lead_stats, wallet_balance, and whoami break the pattern slightly, but overall naming is highly predictable.
309 tools is an extreme count for any MCP server, even a broad platform. This creates significant cognitive load and navigation overhead for agents, and far exceeds the well-scoped 3-15 tool range where coherence is strongest.
The tool surface is remarkably comprehensive across bots, contacts, campaigns, flows, knowledge bases, personas, marketplace, wallet, and products. Minor gaps exist — lead sources lack update/delete tools, and there is no single get_task or get_webhook alongside their list/update/delete counterparts — but these are workable gaps rather than dead ends.