M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
A

AssetOpsBench

Community
A multi-agent MCP benchmark and framework for industrial asset operations
How to install ↓ Compare GitHub source repository ↗
Category
Other #179 of 230
Stars
★ 2.4k Very popular
Runtime
Python
Credentials
No credential needed
License
Apache-2.0
Last commit
Tools
38
45FMRS · D

IBM/AssetOpsBench is an Apache-2.0 industrial asset-operations AI-agent benchmark and framework with IoT, FMSR, TSFM, work-order, and vibration tools plus multiple agent-orchestration approaches. The material confirms MCP support but does not provide a server manifest, transport protocol, authentication method, or client configuration.

Strongest · Maintenance 14/20 Weakest · Reliability 6/20

Reliability
6/20
Security and permissions
7/20
Maintenance
14/20
Documentation
10/20
Setup experience
8/20
Why each score
Reliability 6/20
The README claims MCP support and lists multiple servers and tools, but no cached manifest, actual server entry point, initialization/tool-list handshake evidence, tests, or CI are provided. The quick-start scenario is also marked “to be enabled.” The materials therefore establish only the intended interface, not clean startup, controllable dependencies, error handling, or behavioral consistency, warranting substantial deductions.
Security and permissions 7/20
The materials provide no evidence for credential handling, authentication, permission boundaries, data isolation, or external network flows. Tools include potentially state-changing or execution-related operations such as add_failure_modes, register_model, run_recipe, run_plan, evaluate, and work-order generation, but no confirmation controls, least-privilege model, or dangerous-operation scope is documented. No listed red-line risk is evidenced, but the missing controls require significant deductions.
Maintenance 14/20
The repository is not archived, uses the Apache-2.0 license, and the README presents IBM maintainers plus ongoing research and competition activity; it also reports 46 open issues. This supports relatively strong maintenance activity. However, the supplied material contains no commit history, formal release process, dependency-update policy, issue-response evidence, or security-response channel, so a near-full score is not justified.
Documentation 10/20
The README supplies project scope, domain servers, some tool names, installation commands, example scenarios, documentation links, and MCP context. It lacks a verifiable manifest, complete parameter and return-value documentation, authentication configuration, mainstream-client connection examples, limits, cost information, troubleshooting, and side-effect disclosures. It also contains inconsistent scenario counts, including 141+ in the overview versus the supplied description’s 460+, so the result is only moderate.
Setup experience 8/20
The repository provides clone, pip install -e ., and a sample run command, with a link to INSTRUCTIONS.md for further setup. However, the server path is the repository root, the manifest is unavailable, and no mainstream MCP client configuration, definitive launch command, platform compatibility, environment-variable requirements, or dependency preparation is supplied. The sample scenario is marked as not yet enabled, making the path to a working connection fragile and unverifiable.

Static review · not runListed 2026-07-29

Read the FMRS scoring method →

Fit and risk

Best for

  • Maintenance engineers
  • Reliability specialists
  • Facility planners
  • Industry 4.0 researchers

Not for

  • Users requiring verified production connectors, authentication, or transport details.
  • Users seeking general-purpose chat or non-industrial asset-maintenance tools.
  • Users expecting to run the complete scenarios without consulting the project setup documentation.

Required permissions

  • Requires access to the simulated industrial environment data used by the project.
  • Depending on the task, it may need to read sensor, asset, FMEA, alert, and work-order records.
  • Specific authentication, authorization, and write permissions are not described in the provided material.

Risks and side effects

  • The server manifest is unavailable, so the actual tool surface, startup method, transport, and authentication details require confirmation.
  • The project is under active development; the main branch may differ from branches used for publications.
  • Operations such as adding failure modes, registering models, and running recipes may change records in the project environment; permissions and rollback behavior are not described.

Setup

Before you start

Runtime：Python

From the repository root, run: git clone https://github.com/IBM/AssetOpsBench.git; cd AssetOpsBench; pip install -e .。MCP servers and the plan-execute runner are located under src/; full setup, MCP server, and plan-execute instructions are provided in INSTRUCTIONS.md. The README marks the example scenario as to be enabled.

Check that it works

After cloning the repository and running pip install -e ., connect the IoT MCP server in your client and confirm tools like sites and asset_ids appear in the tool list, then ask "List all assets in MAIN site" to verify the connection.

Troubleshooting

  1. Confirm that IBM/AssetOpsBench is cloned and run pip install -e . from the repository root.
  2. Check that the Python environment and dependencies are installed.
  3. Read INSTRUCTIONS.md for MCP server and plan-execute runner configuration.
  4. Verify that the selected branch contains the required code; the README identifies main as the active development branch and names separate publication branches.
  5. If the example scenario fails, note that the README labels it as to be enabled.

Things to try

Once connected, you can ask your AI assistant things like:

  • List all assets in MAIN site
  • List known failure modes for asset class pump
  • Find a forecasting model for Chiller 6 and check it can be loaded
  • Generate a work order for Chiller 6 anomaly detection

Tools 38

sites read-only
Lists sites.
asset_ids read-only
Lists asset identifiers.
asset_detail read-only
Gets asset details.
assets read-only
Lists assets.
find_assets_by_sensors read-only
Finds assets by sensors.
installed_sensors read-only
Lists installed sensors.
measured_sensors read-only
Lists sensors with measured data.
get_failure_modes read-only
Gets failure modes.
Show 30 more tools
generate_failure_modes read-only
Generates failure modes.
add_failure_modes writes
Adds failure modes.
generate_failure_mode_sensor_mapping read-only
Generates failure-mode-to-sensor mappings.
list_tasks read-only
Lists time-series tasks.
profile_series read-only
Profiles time series.
characterize_series read-only
Characterizes time series.
data_quality read-only
Checks data quality.
list_models read-only
Lists models.
search_models read-only
Searches models.
find_models read-only
Finds models.
resolve_model read-only
Resolves a model.
model_template read-only
Provides a model template.
register_model writes
Registers a model.
register_finetuned writes
Registers a fine-tuned model.
hf_stats read-only
Gets Hugging Face statistics.
list_features read-only
Lists features.
search_features read-only
Searches features.
extract_features read-only
Extracts features.
select_features read-only
Selects features.
recipe_template read-only
Provides a recipe template.
run_recipe writes
Runs a recipe.
run_tabular_recipe writes
Runs a tabular recipe.
run_plan writes
Runs a plan.
evaluate writes
Evaluates run results.
list_runs read-only
Lists runs.
list_results read-only
Lists results.
get_work_order_distribution read-only
Gets work-order distributions.
predict_next_work_order read-only
Predicts the next work order.
compute_fft_spectrum read-only
Computes an FFT spectrum.
compute_envelope_spectrum read-only
Computes an envelope spectrum.

Use cases

Query industrial sites, assets, and sensor information.
Analyze failure modes, sensor relationships, and time-series data.
Find, run, and evaluate time-series models and feature recipes.
Explore multi-agent orchestration and evaluation for industrial asset maintenance.

Overview

AssetOpsBench is an open framework for building, orchestrating, and evaluating domain-specific AI agents for Industry 4.0 asset operations and maintenance. It provides industrial scenarios, domain MCP servers, agent frameworks, and evaluation workflows covering IoT, failure modes and sensor relationships, time-series foundation models, work orders, and vibration analysis.

Similar servers

BioMCP 75 · B

One binary. One grammar. Evidence from the biomedical sources you already trust.

★ 659 · Tools 11 Compare with this →
Compare with other MCP servers

Source revision e844f6468c6c Data synced 2026-10-19 Read the FMRS scoring method