The request is for a task-estimator, a tool that estimates how long or how costly a task is. None of the candidate sources is a dedicated task-estimation product, so the closest fits are the ones that expose a task or benchmark/evaluation operation. Azure's agentops evaluation run 8, Microsoft's benchmark operation that returns a job to poll 2, and Open-RMF's estimate_robot_task_request for estimating a robot task 10 are the most relevant, though each covers only part of the request.
A question back: Which kind of task do you want estimated: a robot task, a model or AI agent evaluation, or a software project's effort?
- 1
- 2
Microsoftmicrosoft.comverified and signed
Its benchmark action benchmarks or evaluates a model and returns a job_id to poll for status 2.
official:MCP server · Playwright MCP ServerSDK · markitdownlibrary · windows-sysCLI · playwright-mcppackage · Azure DevOps (ADO)+151 community tools built for it
for agents: Tasks · Tasks · benchmark
- 3
Azureazure.comfound in code, not yet checked
Its agentops_eval_run action runs an AgentOps evaluation and returns a summary 8.
official:MCP server · mcp-kubernetesSDK · azure-corelibrary · Azure.CoreCLI · azcopypackage · Containerization Assist+287 community tools built for it
for agents: agentops_eval_run · scan · scan-carbon-emissions
Sources
- 2Microsoft – AI, Cloud, Productivity, Computing, Gaming & Appsmicrosoft.com · read 2026-10-29
- 8Cloud Computing Services | Microsoft Azureazure.microsoft.com · read 2026-10-29
- 10Open-RMFopen-rmf.org · read 2026-10-29
For agents: the 10 actions behind this answer
body-pose-estimation · body-pose verified and signed
Task: Body Pose Estimation
Tasks · Microsoft.SourceLink.Common verified and signed
arize-evaluator · arize-evaluator verified and signed
Handles LLM-as-judge evaluation workflows on Arize including creating/updating evaluators, running evaluations on spans or experiments, managing tasks, trigger-run operations, column mapping, and continuous monitoring. Use when the user mentions create evaluator, LLM judge, hallucination, faithfulne
create_task · supabase found in code, not yet checked
Create an open task for yourself and return it. Each call creates a new task; do not retry blindly after a connection failure.
AI · sagemaker verified and signed
data-profilers · pydeequ verified and signed
radar-get-ai-inference-summary-by-task · Cloudflare API verified and signed
Get Workers AI tasks summary
agentops_eval_run · agentops found in code, not yet checked
Run an AgentOps evaluation and return a summary.
task-scheduling · dask verified and signed
estimate_robot_task_request · Open-RMF fleet API (multi-fleet robot orchestration) verified and signed
Request to estimate a task for specific robot
Labels are the instruments' verdict and are never changed by ranking. Agents and robots get this same answer as JSON at /api/v1/find?q=…&answer=1 and through the MCP server.