Skip to main content
Glama

Duvo MCP

by io.github.duvoai

Start Run

startRun

Start a new agent run. Returns immediately with run info - does not wait for completion.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
messageNoOptional initial message to start the run with (e.g. a trigger payload or user instruction)
team_idNoDuvo team UUID to operate on. API keys are pinned to a single team — omit this (it falls back to the key's team) or pass that same team; a different team is rejected. OAuth callers, who can span multiple teams, should pass the target team here.
agent_idYesThe agent ID to run
sandbox_idNoOptional sandbox ID with pre-uploaded files
webhook_urlNoWebhook URL to POST events to (human_request, run_completed, run_failed, run_interrupted)

Schema Changelog

Changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. First observed

TDQS

A4.2/5.0
Behavior4/5

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

Annotations only indicate this is non-read-only, non-idempotent, and not destructive. The description adds a meaningful behavioral fact beyond those: the run starts and the call returns immediately without waiting for completion. This is valuable because it sets agent expectations about the tool's asynchronous nature.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two short sentences, zero wasted words. The core action is front-loaded, and the key behavioral caveat ('does not wait for completion') is placed immediately after the action statement.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The description covers the essential async contract and the schema documents all parameters, including webhook event types. It does not specify the exact shape of 'run info' or how to retrieve the run later, but sibling tools like getRun and the webhook_url parameter fill most of that gap.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so all five parameters are already documented with meaningful descriptions, including the nuanced team_id behavior. The tool description itself adds no parameter-specific meaning, so the baseline score of 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses a specific verb and resource ('Start a new agent run') and adds a distinguishing async behavior ('Returns immediately with run info - does not wait for completion'). This clearly separates it from sibling tools like getRun, listRuns, and stopRun without needing to open the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It clearly states when to use the tool: when starting a new agent run. The async disclosure implies you should not expect completion results synchronously, giving useful context. However, it does not explicitly name alternatives like getRun for polling status or stopRun for cancellation.

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

All Duvo MCP toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

B3.1/5.0
Disambiguation2/5

Despite detailed descriptions, many tool names are highly ambiguous, with multiple tools covering the same conceptual actions (e.g., acceptClarityCaptureSuggestion vs. acceptClarityTeamAssignmentSuggestion, or the many deleteClarity*Interview tools). The set is so large that distinguishing between, say, listClarityFolders, listClarityProcesses, and listClarityProcessSummaries requires reading deep into descriptions, reducing agent selection accuracy.

Naming Consistency4/5

The naming convention is predominantly verb_noun (e.g., createClarityProcess, listAgents, deleteQueue), and is remarkably consistent across the 316 tools. There are only minor deviations, such as 'fileSuggestedClarityProcesses' (verb + adjective noun) and 'bulkUpdateCasePriority' (where 'bulk' could be seen as a prefix), but overall the pattern holds strongly.

Tool Count1/5

With 316 tools, this server is extremely oversized for any single agent to manage effectively. The massive number of tools suggests poor modularization—many of these tools likely belong in separate, smaller servers focused on specific domains (e.g., Clarity, Pulse, Agent management). The cognitive load for an agent to choose from 316 options is very high, leading to frequent misselection.

Completeness4/5

The tool surface covers an extraordinarily wide range of operations across the Duvo platform: agents, runs, cases, queues, Clarity processes, skills, integrations, notifications, teams, and more. Most resource types have full CRUD and lifecycle management. Notable minor gaps exist (e.g., no tools for managing specific notification batch severities dynamically, and some interview management is missing batch operations), but for the platform's scope, coverage is impressively thorough.

Resources