MCP-OpenStack-Ops
Read-only, project-scoped OpenStack inspection and monitoring server; the supplied schema exposes only get/search-style tools, with no create/update/delete tools registered.
Inspect compute: instances by name/ID/status/search with pagination, server events, server groups, attached volumes, availability zones, hypervisors, keypairs, and service status.
Inspect networking: networks/subnets, routers, floating IPs and pools, security groups, QoS policies, plus opt-in cross-project reads.
Inspect block storage: volumes, snapshots, backups, volume types, generic groups, transfers, QoS specs, and server volumes.
Inspect images and image membership.
Inspect identity and quotas: projects, domains, groups, users, role assignments, quotas, usage statistics, and resource monitoring allocations/capacity.
Inspect Heat stacks and details: resources, events, templates, outputs, and nested depth.
Inspect Octavia load balancers: details, listeners, pools, members, health monitors, L7 policies/rules, amphorae, providers, flavors, availability zones, and quotas.
Inspect Barbican secret metadata: secrets, containers, orders, ACLs, quotas, and secret stores without returning payloads.
Query Gnocchi telemetry: resources, metrics, measures, aggregates, and service info for measured utilization and trends.
Use read-only safety controls: tools are scoped to the current project, and cross-project reads are opt-in and policy-dependent.
Provides comprehensive OpenStack project management and monitoring capabilities, including compute, networking, storage, images, identity, Heat, and Octavia load balancing operations, all scoped to a single project with safety controls.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MCP-OpenStack-Ops list all running instances in my project"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
English | Korean
MCP-OpenStack-Ops
MCP OpenStack Operations Server: A comprehensive MCP (Model Context Protocol) server providing OpenStack project management and monitoring capabilities with built-in safety controls and single-project scope.
What is MCP-OpenStack-Ops?
MCP-OpenStack-Ops lets an AI assistant such as Claude or a model in OpenWebUI inspect and operate an OpenStack project in plain language. It is an MCP server that turns requests like "which instances are stopped?" into OpenStack API calls and returns the results.
Read-only by default: tools that create, change or delete resources are not even registered unless you enable them.
One project at a time: each server works inside one OpenStack project and checks resource ownership.
Broad coverage: 121 tools (61 read-only, 60 management) for Nova, Neutron, Cinder, Glance, Keystone, Heat, Octavia, Barbican, and Gnocchi/Aodh telemetry.
For an AI-generated walkthrough of the code, see
.
Related MCP server: mcpdeployment
What can you do with it?
Showcase: a Heat autoscaling stack created from natural language alone

No Heat template was written by hand: the stack was created from this two-line natural-language request.
Use the test-rockylinux-9 VM (web server) as the source for a new base image. Create and bring up a Heat stack with a TLS-terminating load balancer and autoscaling web servers.
- Run 2 web servers under normal conditions, with a maximum of 4. Scale out by 1 when average CPU utilization stays above 60% for 5 minutes, and scale in by 1 when it stays below 40% for 5 minutes, down to a minimum of 2. Use a 900-second cooldown and set the stack's timeout_mins to 60.Using this server's tools, the MCP client snapshotted a web server VM into a base image, generated a self-signed certificate and stored it in Barbican, and created a Heat stack with an Octavia load balancer that terminates HTTPS and an autoscaling group driven by Aodh CPU alarms. Under CPU load the group grew from two to four web servers and shrank back to two when the load ended. The request, templates and a walkthrough of the chart are in the Heat autoscaling tutorial.
Then ask how it is doing, in one line

A one-line request for the members' CPU over the last hour, at 1-minute intervals, returned per-member data from Gnocchi through this server's tools; the client drew the chart, summary and CSV. It shows the group mean above 60% for five minutes and the third member added right after. (The session was in Korean; the screenshot shows it translated into English.)
Ask questions and give instructions in natural language, for example:
Show me the overall cluster status
Find instances in SHUTOFF state
Show all networks and their subnets
List all volumes and their status
Show the vCPUs and memory allocated to my project and the remaining quota
Show CPU usage of web-01 over the last hour
Stop all instances with name containing 'test'The last request changes resources, so it works only after you enable modification tools. Telemetry requests need Ceilometer and Gnocchi in the cloud.
More requests to try: Example queries by area lists copy-ready requests, with the tools each one uses, for cluster status, instances, networks, storage, images, identity, Heat, load balancers, Barbican secrets and TLS listeners, allocation and quotas, Gnocchi telemetry, Aodh alarms, troubleshooting, security, batch actions, and multi-tool reports.
Example: cluster status in Claude Desktop

The same project in the OpenStack Dashboard (Epoxy 2025.1)

Quick Start
You need an OpenStack account (Keystone host, project, user name, password) whose APIs are reachable from the computer that runs the server. Pick one of the two paths below; both start read-only.
Option A: Claude Desktop with the extension bundle (fastest)
Download
mcp-openstack-ops-<version>.mcpbfrom the latest GitHub Release.In Claude Desktop, open Settings > Extensions > Advanced settings > Install Extension⦠and select the file.
Fill in the settings form: Keystone host, project, user name and password are required; ports, domains and region are prefilled. Leave Allow modify operations off.
Ask Claude:
List the instances in my OpenStack project.
Other Claude Desktop setups (uvx, a source checkout, the Docker server
through mcp-remote, a remote connector) are in the
Claude Desktop guide.
Option B: OpenWebUI with Docker Compose
Requirements: Git, Docker with the Compose plugin, and a model provider for OpenWebUI whose model supports tool calling.
1. Configure
git clone https://github.com/call518/MCP-OpenStack-Ops.git
cd MCP-OpenStack-Ops
cp .env.example .env
cp mcp-config.json.http mcp-config.json.http.local
chmod 600 mcp-config.json.http.local
openssl rand -hex 32 # use the output as WEBUI_SECRET_KEY
vi .env # set WEBUI_SECRET_KEY and the OS_* values for your cloud2. Start
docker compose up -d
docker compose ps # wait until mcp-server is healthyThis starts OpenWebUI (http://localhost:3000), the MCP server
(http://localhost:8000/mcp) and MCPO (http://localhost:8080).
3. Connect a model provider
These steps apply to both the direct MCP connection and MCPO.
Open
http://localhost:3000and create the administrator account.Under Settings β Admin β Connections, add an OpenAI connection with URL
https://api.openai.com/v1, AuthBearerwith your OpenAI API key, and API Type set to Chat Completions, then save it. For another provider, see the OpenWebUI guide.The Quick Start calls tools through Chat Completions with model reasoning turned off, using this default already present in
.env.example:DEFAULT_MODEL_PARAMS={"reasoning_effort":"none"}This value requires API Type: Chat Completions. With Responses, OpenWebUI 0.13.0 still sends
reasoning_effort, and every request fails withUnsupported parameter: 'reasoning_effort'. If theopen-webui-datavolume already exists, set the same value under Settings β Admin β Models β Model Defaults β Model Parameters because its saved database value takes precedence. To keep model reasoning through Responses instead, see the OpenWebUI guide.
4. Connect the MCP server
Under Settings β Admin β Tools β Integrations β External Tool Servers, add a connection of type
MCP (Streamable HTTP)with URLhttp://mcp-server:8000/mcpand authenticationNone.Under Settings β Admin β System β Interface β Chat, turn on Tool Search and Defer Built-in Tools, and set Deferral Threshold to
0so that the full tool list fits the model's limits.In a new chat, enable the connection under Integrations β Tools and ask:
Use mcp-openstack-ops to list the instances in my OpenStack project.
MCPO is optional. MCPO is a proxy that exposes the MCP tools as an OpenAPI (REST) server; OpenWebUI does not need it to use this server. The Compose file starts it so that you can choose either connection. To run without it, start only
docker compose up -d mcp-server open-webui.
The OpenWebUI guide explains the direct and MCPO connections, every OpenWebUI setting, OpenAI reasoning models, and troubleshooting.
Other ways to install
Setup | Guide |
Claude Desktop with | |
Claude Code | |
OpenWebUI through MCPO (OpenAPI) | |
Each user signs in with their own OpenStack Application Credential (experimental) | |
Several projects, HTTPS and private CAs, custom endpoints and ports |
To let the assistant create, change or delete resources, set
ALLOW_MODIFY_OPERATIONS=true (or turn on Allow modify operations in the
Claude Desktop form) and restart the server. OpenStack policy still decides
what the account may do.
Features
β Project-Scoped Operations: Every tool enforces the configured
OS_PROJECT_NAME, validating resource ownership so actions stay inside a single tenant.β Safety-Gated Writes: Modify (
set_*) tooling only registers whenALLOW_MODIFY_OPERATIONS=true, keeping default deployments read-only and auditable.β Broad Tool Coverage: 121 tools across compute, networking, storage, images, identity, Heat, Octavia load balancing, Barbican secrets and telemetry. Project resources are scoped to the active project; cloud-wide administrative APIs remain subject to OpenStack policy.
β Bulk & Filtered Actions: Instance, volume, network, image, snapshot, and keypair managers accept comma-delimited targets or filter criteria to orchestrate bulk changes intentionally.
β Post-Action Feedback & Async Guidance: Management tools report per-resource results, the status read back after the action, and timing notes for asynchronous operations.
β Monitoring & Usage Insights:
get_service_status,get_resource_monitoring,get_usage_statistics, and quota tools report service probes, project allocations, separate quota limits, and API-backed compute allocation-hours. These tools do not collect live utilization or trends; read-onlyget_gnocchi_*tools return telemetry time series when Ceilometer and Gnocchi are deployed.β Unified Instance Queries: The
get_instancetool consolidates name, ID, status, and free-form search paths with pagination plus summary/detailed modes.β Server Insight & Audit Trail: Dedicated tools expose server events, hypervisor details, availability zones, quotas, and resource ownership to speed diagnostics.
β Load Balancer Management: Octavia tools cover listeners, pools, members, health monitors, flavors, quotas, and amphora operations with the same safety gates.
TERMINATED_HTTPSlisteners take certificates stored in Barbican, and pools support non-terminated HTTPS (TLS passthrough) and backend re-encryption, including client authentication (mTLS).β Barbican Secrets Without Payloads: Barbican tools manage secret metadata, containers, key-generation orders, ACLs and self-signed TLS certificates. Secret payloads are never accepted or returned, so key material cannot reach the LLM; register your own certificates outside MCP (Barbican and TLS guide).
β Connection & Deployment Flexibility: Connection caching, configurable service endpoints, Docker packaging, a Claude Desktop extension, and both
stdio/streamable-httptransports support proxy/bastion and multi-project setups.π§ Advanced: Cross-Project Reads: Administrators can ask about every project at once ("List VMs per project"), and operators can give read-only, cloud-wide views to non-admin teams through OpenStack policy. Off by default; see the cross-project reads guide.
Safety and security
Read-only by default. Management tools are registered only with
ALLOW_MODIFY_OPERATIONS=true.One project per server. Resources are filtered by the current project and checked for ownership; run one server per project to manage several. Reading across projects is an advanced, opt-in feature; see the cross-project reads guide.
Protect HTTP access. Over
streamable-http, enable Bearer token authentication (REMOTE_AUTH_ENABLE=true) and HTTPS before exposing the server beyond localhost.
Details: security guide and multi-user guide.
Authentication support by client
Each server authenticates to OpenStack with either one account's password (the
default) or per-user Application Credentials (multi-user mode, streamable-http
only). Support differs by client; for example, the Claude Desktop extension
bundle supports only the password, while Claude Desktop with mcp-remote in
claude_desktop_config.json supports both. See
Authentication support by client.
Compatibility
β οΈ Compatibility Notice: This MCP server is developed and optimized for OpenStack Epoxy (2025.1) as the primary target environment. However, it is compatible with most modern OpenStack releases (Dalmatian, Caracal, Bobcat, etc.) as the majority of APIs remain consistent across versions. Only a few specific API endpoints may require adaptation for full compatibility with older releases.
π§ Coming Soon: Dynamic multi-version OpenStack API compatibility is actively under development and will be available in upcoming releases, providing seamless support for all major OpenStack deployments automatically.
Officially Supported Releases:
β OpenStack Epoxy (2025.1) - Fully tested
β OpenStack Dalmatian (2024.2) - Fully tested
OpenStackSDK Version Reference:
OpenStack Release | Recommended SDK Version | Notes |
Epoxy (2025.1) |
| Current default range; |
Dalmatian (2024.2) |
| Fully compatible |
Caracal (2024.1) |
| May require testing |
Bobcat (2023.2) |
| May require testing |
Wallaby (2021.1) |
| Downgrade required |
To change the SDK version for an older release, see OpenStack SDK version for older releases.
Documentation
Guide | Contents |
Example queries by area, bulk and filtered actions, post-action status, pagination | |
Docker Compose stack, direct MCP versus MCPO, OpenWebUI settings, troubleshooting | |
Extension bundle, | |
Registering the HTTP server in Claude Code | |
Command-line options, environment variables, access modes and | |
Authentication support by client, project isolation, read-only default, Bearer authentication | |
Per-user Application Credential authentication (experimental) | |
Advanced, opt-in: reading across projects, setups, required OpenStack policy | |
Barbican prerequisites, why secret payloads are excluded, self-signed certificates, | |
Heat stack with a | |
Tests, live tool checks, adding tools, contributing | |
Outstanding implementation and live-validation work | |
Every tool, its CLI equivalent, and its validation status | |
Dates, revisions, conditions and counts of the live and unit-test checks |
π OpenStack CLI vs MCP Tools Mapping
β denotes an implemented API path, not certification of every action or OpenStack deployment. β οΈ Live unverified identifies paths without successful live validation; β Not implemented identifies unavailable capabilities. Service availability, permissions, and asynchronous completion affect results. See API prerequisites and result semantics and the multi-user setup guide.
Validation evidence: the live checks, Claude Code prompt checks and unit-test results behind the Status values below are recorded in docs/validation.md.
Detailed Mapping by Category
1. π₯οΈ Compute (Nova)
OpenStack CLI Command | MCP Tool | Status | Notes |
|
| β | Separate reader projects each returned exactly ten owned VMs; bounded lab validation only. One call returns at most |
|
| β | Own fixture queries completed; foreign private ID returned no details; no 16-owned-ID batch claim. Each instance reports |
|
| β | ENHANCED - Bulk creation support. Boots the image on the flavor's local disk unless a volume option below is given. 2026-11-10 targeted check: a server reached ACTIVE |
|
| β | Root disk on a new Cinder volume created from the image; optional |
|
| β | Boots one server from an existing available, bootable volume, without an image. 2026-11-10 OpenStack 2026.1 check (3.7.6): the server reached ACTIVE on that volume; an in-use, non-bootable or missing volume, and two server names, were refused without creating a server |
|
| β | Tool schema exposes all actions and |
|
| β | Tool schema exposes delete and optional force deletion; bulk deletion with |
|
| β | Nova backup action with rotation; asynchronous, no custom metadata |
|
| β | Accepts optional |
|
| β | Actions are exposed in the tool schema; instance shelving was not live-retested in this change |
|
| β | Instance locking. |
|
| β | Uses the Nova remote-consoles API (the legacy console actions return 404 on current Nova). |
|
| β | Instance pausing |
|
| β | Instance suspension |
|
| β | Uses the existing |
|
| β | Resize confirmation |
|
| β | Resize revert |
|
| β | Uses |
|
| β | Tool schema exposes both actions and optional |
|
| β | Live migration |
|
| β οΈ Live unverified | Server evacuation. |
|
| β | Migration listing. With |
|
| β | Migration details |
|
| β | Migration abort |
|
| β | Migration confirmation |
|
| β | Force migration completion |
|
| β | Network attachment. Nova creates the interface port, so removing the network deletes it again. 2026-11-10 targeted check: attached |
|
| β | Network detachment. 2026-11-10 targeted check: no interface port was left, and the network could be deleted afterwards |
|
| β | Port attachment by port name or ID (a name was previously reported as not found). 2026-11-10 targeted check: attached by name |
|
| β | Port detachment by name or ID; detaching does not delete a port you created. 2026-11-10 targeted check: detached by name and the port was kept |
|
| β | Floating IP association |
|
| β | Floating IP disassociation |
|
| β | Fixed IP addition through a new interface whose port Nova creates. 2026-11-10 targeted check: an address from the requested network was assigned |
|
| β | Detaches the whole interface that holds the address, which can be the primary interface. 2026-11-10 targeted check: removed and no interface port was left |
|
| β | Security group addition |
|
| β | Security group removal |
|
| β | Member attach reached in-use via set_server_volume; guest I/O not tested. The multiattach precheck reads the SDK |
|
| β | Member detach reached available via set_server_volume. 2026-11-10 targeted check (administrator): reached available |
|
| β | Server property setting |
|
| β | Server property unsetting |
|
| β Not implemented | Nova's trigger-crash-dump action (microversion 2.17, openstacksdk |
|
| β | Pass the server name or UUID via |
|
| β | Server group listing. Reads |
|
| β | Server group management. 2026-11-10 targeted check: create and show reported the policy and members, then the group was deleted |
|
| β | Requires modification tools to be enabled for registration |
|
| β | Flavor management |
|
| β | Empty scoped read completed; member keypair creation/deletion read-back tested separately |
|
| β | Explicit member MCP create/delete read-back passed; no successful Claude write workflow claimed |
|
| β | Administrator lookup verified; missing per-host capacity and incomplete totals are null, not zero. Global enhanced statistics remain separate; when |
|
| β | Compute/Cinder availability reads the SDK state mapping; missing availability is null. Actual available zones verified |
2. π Network (Neutron)
OpenStack CLI Command | MCP Tool | Status | Notes |
|
| β | Two owned private networks plus one policy-visible shared external network per reader; no other test project private networks |
|
| β | Foreign-ID negative checked. A name that matches nothing returns an explicit error (previously an empty success). 2026-11-10 targeted check (administrator): named lookup returned the network and a missing name returned the error; a dual-project positive lookup by reader roles is still not asserted |
|
| β | Default mtu=0 omits the wire field: backend MTU 1450 verified; explicit 1400 retained in both member projects |
|
| β | Both default/explicit-MTU test networks removed through actual TCP MCP, API absence confirmed |
|
| β | ENHANCED - Bulk updates |
|
| β | Owned and shared external network subnets returned; three visible subnets per test reader |
|
| β οΈ Live unverified | Create omits an unset |
|
| β οΈ Live unverified | Only part of composite error requests in this round; no independent reader result attribution |
|
| β | Project-owned routers; attached interfaces are not removed automatically |
|
| β | Floating IP listing. A failed query returns an error instead of a placeholder address. 2026-11-10 targeted check: listed |
|
| β | Floating IP creation. |
|
| β | Floating IP deletion. 2026-11-10 targeted check: released |
|
| β | Floating IP property setting |
|
| β | Floating IP details. 2026-11-10 targeted check: shown by ID |
|
| β | Detaches the port and keeps the floating IP. |
| β | β Not implemented | No tool clears a floating IP's QoS policy or tags. |
|
| β | Available global pool capacity remains unknown ( |
|
| β οΈ Live unverified | Implemented; test platform port-forwarding endpoint returns HTTP 404 |
|
| β οΈ Live unverified | Implemented; test platform port-forwarding endpoint returns HTTP 404 |
|
| β οΈ Live unverified | Implemented; test platform port-forwarding endpoint returns HTTP 404 |
|
| β οΈ Live unverified | Implemented; test platform port-forwarding endpoint returns HTTP 404 |
|
| β οΈ Live unverified | Implemented; test platform port-forwarding endpoint returns HTTP 404 |
|
| β | Scoped reader list completed; default and test groups visible. A failed query returns an error instead of a placeholder |
|
| β | Project-owned security groups |
|
| β οΈ Live unverified | Implemented for project-owned groups. Create requires an explicit CIDR or project-owned remote group; delete verifies the rule belongs to the selected group. Not live-tested in this change |
|
| β οΈ Live unverified | Implemented list action requires modification-tool registration; not exercised in this round |
|
| β οΈ Live unverified | Create omits unset |
|
| β οΈ Live unverified | Implemented; test platform QoS endpoint returns HTTP 404 |
|
| β οΈ Live unverified | Implemented; test platform QoS endpoint returns HTTP 404. Policy containers only; no bandwidth rules |
|
| β οΈ Live unverified | Reader service-policy restrictions; partial data is not proof of global health. 2026-11-10 targeted check returned agents as administrator only |
|
| β οΈ Live unverified | Enable/disable cloud-wide agents; usually admin-only. The live tool check called only action="list"; enable and disable were not exercised |
3. πΎ Storage (Cinder)
When the Block Storage service is not deployed, the Cinder volume, snapshot, type, backup, group, QoS, transfer and server-volume paths report an explicit "Block Storage (Cinder) service is not available" error with endpoint URLs redacted. This was checked live on OpenStack 2026.1 without Cinder with release 3.7.3 (see validation); the storage code is unchanged since.
OpenStack CLI Command | MCP Tool | Status | Notes |
|
| β | Exactly ten owned volumes per reader; no other test project volume IDs returned. |
|
| β | Listing-based client filtering; no dedicated get_volume_list filter argument |
|
| β | Volume creation/deletion. 2026-11-10 targeted check: a 1 GB volume reached available and two volumes were deleted |
| β | β Not implemented |
|
|
| β | Positive new_size takes precedence over size; size remains a fallback. A test volume reached the requested size and available state; API acceptance alone is not completion |
| β | β Not implemented | set_volume has no modify/update action; do not send this unsupported action |
|
| β | Volume type listing. A failed query returns an error instead of a placeholder |
|
| β | Volume type management; usually admin-only |
|
| β | Two owned base snapshots per reader; API final states checked |
|
| β | Explicit member MCP snapshots reached available and were deleted. Post-action status resolves snapshots instead of reporting an unknown resource type; no restore/I/O claim. 2026-11-10 targeted check (administrator): reached available and was deleted |
|
| β | Three owned available backups per reader; base backup creation used real Cinder APIs |
|
| β | Volume backup management. 2026-11-10 targeted check: a backup reached available and was deleted |
|
| β | Restores to a new volume named by |
|
| β | Request listing only; authorization keys are excluded |
|
| β | Server volume listing. A server that does not exist returns an explicit error (previously an empty success). 2026-11-10 targeted check: listed the attached volume; a missing server returned the error |
|
| β | Both paths reached in-use then available in both member projects; foreign endpoints rejected; guest I/O not tested |
|
| β | Requires Cinder generic groups (API >= 3.13) |
|
| β | Create/list/show/delete; create requires group_type and volume_types |
|
| β | Cinder QoS specifications; subject to cloud policy |
|
| β οΈ Live unverified | Shows one specification and its current volume type associations. Added 2026-11-17; unit tests only |
|
| β | Create/list/show/set/delete specifications. Creation does not automatically associate a volume type; use the explicit |
|
| β οΈ Live unverified | Associates a specification with a volume type by name or ID, then reads the associations back and reports |
|
| β οΈ Live unverified | Disassociates one volume type by name or ID and reads the associations back. Added 2026-11-17; unit tests only |
|
| β οΈ Live unverified | Disassociates every volume type and reads the associations back. Added 2026-11-17; unit tests only |
4. πΌοΈ Image (Glance)
OpenStack CLI Command | MCP Tool | Status | Notes |
|
| β | Image listing. A failed query returns an error instead of an empty list. 2026-11-10 targeted check: listed |
|
| β | Specific image query |
|
| β | Creates the image record with min_disk, min_ram and properties only. No image data is uploaded, so the image stays |
| β | β Not implemented | No tool uploads or imports image data |
|
| β οΈ Live unverified | Only images owned by the current project; other projects' public, shared or community images are refused (previously allowed). This 2026-11-17 change is unit-tested only |
|
| β οΈ Live unverified | Only images owned by the current project (previously any visible image). This 2026-11-17 change is unit-tested only |
| β | β Not implemented |
|
|
| β | Add a project to an image with shared visibility |
|
| β | Image member listing |
|
| β | Image member management |
|
| β οΈ Live unverified |
|
|
| β οΈ Live unverified |
|
5. π₯ Identity (Keystone)
OpenStack CLI Command | MCP Tool | Status | Notes |
|
| β οΈ Live unverified | Reader denied by Keystone; explicit MCP failure observed; successful listing not validated in this round |
|
| β οΈ Live unverified | Depends on policy-limited user listing; not independently validated for the new readers |
|
| β | Password from a server environment variable; no role grants |
|
| β | Without |
|
| β | Without |
|
| β | Project management |
| β | β Not implemented |
|
|
| β | Role listing |
|
| β | A failed query returns an error instead of a placeholder assignment. 2026-11-10 targeted check: listed as administrator; reader roles were denied by Keystone policy in an earlier round |
|
| β | Role creation; |
| β | β Not implemented |
|
|
| β | Subject to Keystone policy |
|
| β | Disable explicitly with update before delete |
|
| β | Subject to Keystone policy |
|
| β | Group management |
|
| β οΈ Live unverified | Reader service-policy restrictions; partial data is not proof of global health. Compute and volume service zones read the SDK |
|
| β | Reads the SDK |
| β | β Not implemented |
|
|
| β οΈ Live unverified | Reader service-policy restrictions; partial data is not proof of global health |
6. π₯ Orchestration (Heat)
OpenStack CLI Command | MCP Tool | Status | Notes |
|
| β | The stack's project ID or Heat self link must match the current project; unknown and foreign owners are excluded. Own and foreign test stacks were checked. |
|
| β Partial | Select the target from the returned list; no name/filter parameter. Resource/event/template/output queries use get_heat_stack_details |
|
| β | Stack creation. 2026-11-10 targeted check: reached CREATE_COMPLETE. |
|
| β | Foreign-project deletion rejected; owned stack deletion completed after the ownership change. 2026-11-10 targeted check: deleted |
|
| β | Requires matching stack ownership. The default |
|
| β οΈ Live unverified |
|
| β | β Not implemented | set_heat_stack supports create, delete, update and validate only |
|
| β | Posts the template with |
|
| β | Actual Heat resources. |
|
| β | Owned stack events retrieved after the ownership change. |
|
| β | Owned stack template retrieved after the ownership change |
|
| β | Owned stack outputs retrieved after the ownership change |
7. βοΈ Load Balancer (Octavia)
The Octavia tools read the openstacksdk attribute names (is_admin_state_up,
is_enabled, the listener load_balancer_id, the L7 rule rule_value and
plural quota fields), list pool members with members(pool), create L7 rules
with create_l7_rule(policy, ...), and filter listeners by load balancer. These
corrections are unit-tested against real openstacksdk resource objects only;
Octavia was not available in any live test cloud.
OpenStack CLI Command | MCP Tool | Status | Notes |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Load balancer listing with pagination; one call returns at most |
|
| β | 2026-11-12 targeted Octavia read check passed. Load balancer detailed information; pools are matched to listeners from each pool's |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Load balancer creation |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Load balancer deletion |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Load balancer property update |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Load balancer statistics |
|
| β | Octavia status tree ( |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Load balancer failover |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Load balancer property unset |
Listener Management | |||
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Listener listing for load balancer; reports |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Listener creation (HTTP/HTTPS/TCP/UDP) |
|
| β | Barbican secret or certificate container by name, ID or href (names resolved to hrefs); |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Listener deletion |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Listener detailed information |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Listener property update |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Listener statistics |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Listener property unset |
Pool Management | |||
|
| β | 2026-11-12 targeted Octavia read check passed. Pool listing (all or by listener); reports |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Pool creation with algorithms |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Pool deletion |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Pool property update |
|
| β οΈ Live unverified | Unit-tested only. Backend re-encryption; certificate references by Barbican name, ID or href. Pool results report these fields. See the Barbican and TLS guide |
|
| β οΈ Live unverified | Unit-tested only. Client authentication for backend re-encryption: the client certificate for members that require mutual TLS (mTLS). openstacksdk 4.22.0 has no such pool field, so a request with it goes to the Octavia API directly; |
|
| β | 2026-11-12 targeted Octavia read check passed. Pool detailed information, including |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
Member Management | |||
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Pool member listing |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Pool member creation |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Pool member deletion |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Pool member property update |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Pool member detailed information |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
Health Monitor Management | |||
|
| β | 2026-11-12 targeted Octavia read check passed. Health monitor listing (all or by pool); reports |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Health monitor creation; types include |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Health monitor deletion |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Health monitor update |
|
| β | 2026-11-12 targeted Octavia read check passed. Health monitor detailed information, including |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
L7 Policy Management | |||
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. L7 policy listing |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. This management tool supports create/delete only |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. L7 policy deletion |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. L7 policy details |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
L7 Rule Management π | |||
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. L7 rule listing |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. L7 rule creation |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. L7 rule deletion |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. L7 rule details |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
Amphora Management π | |||
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Amphora listing |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Amphora details |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Amphora configuration |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Amphora failover |
| N/A | β | Not supported by OpenStack SDK |
| N/A | β | Not supported by OpenStack SDK |
Provider Management | |||
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Provider listing |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
Availability Zone Management π | |||
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Availability zone listing |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Availability zone details |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Availability zone creation |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Availability zone deletion |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
Flavor Management π | |||
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Flavor listing |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Flavor details |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Flavor creation |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Flavor deletion |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
Flavor Profile Management | |||
| N/A | β Not implemented | No registered tool/action for this CLI operation |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
| N/A | β Not implemented | No registered tool/action for this CLI operation |
Quota Management π | |||
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Quota listing |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Quota details |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Quota setting |
|
| β οΈ Live unverified | Implemented; Octavia unavailable in the test platform. Quota reset |
8. π Key Manager (Barbican)
Secret payloads are excluded on purpose. Barbican and openstacksdk can store
and return them, but tool arguments and results pass through the LLM, so no
tool accepts or returns a payload: register your own certificates and keys
outside MCP and reference them by name or href. Reads use Barbican's metadata
API, never the SDK's payload-downloading get_secret/find_secret. See the
Barbican and TLS guide.
OpenStack CLI Command | MCP Tool | Status | Notes |
|
| β οΈ Live unverified | Unit-tested only. Metadata of current-project secrets; |
|
| β οΈ Live unverified | Unit-tested only. By name, ID or href; metadata and consumers only |
| N/A | β Not implemented | Intentionally excluded: payloads would reach the LLM |
| N/A | β Not implemented | Intentionally excluded: payloads would reach the LLM. Register user certificates outside MCP |
|
| β | Key and certificate generated in the MCP server and stored as an unencrypted PKCS#12 |
|
| β οΈ Live unverified | Unit-tested only. By name, ID or href |
|
| β οΈ Live unverified | Unit-tested only. List, or one container by name, ID or href |
|
| β οΈ Live unverified | Unit-tested only. |
|
| β οΈ Live unverified | Unit-tested only. The secrets are kept |
|
| β οΈ Live unverified | Unit-tested only. Raw order data, including |
|
| β οΈ Live unverified | Unit-tested only. |
|
| β οΈ Live unverified | Unit-tested only |
|
| β οΈ Live unverified | Unit-tested only. Secret or container |
|
| β οΈ Live unverified | Unit-tested only. Secret or container; |
Barbican effective quotas ( |
| β οΈ Live unverified | Unit-tested only. Current project's effective quotas. Project quota changes are not implemented |
Secret store backends ( |
| β οΈ Live unverified | Unit-tested only. Served only when Barbican enables multiple secret store backends |
9. π Monitoring & Logging
Allocation and measured telemetry are separate. get_resource_monitoring,
get_quota, and get_usage_statistics report flavor allocations, quotas,
capacity, and allocation-hours from Nova. get_gnocchi_* tools report observed
time series collected by Ceilometer into Gnocchi. Tool descriptions cross-reference
each other so that requests for utilization, load, or trends use Gnocchi. They
also instruct the model to report an uncollected measured metric as not
collected instead of substituting allocations. When the Keystone catalog has no
metric endpoint, the get_gnocchi_* tools report "Metric service (Gnocchi) is
not available in this cloud" (checked live on OpenStack 2026.1).
OpenStack CLI Command | MCP Tool | Status | Notes |
Resource monitoring |
| β οΈ Live unverified | Reader physical-capacity policy restrictions; allocations are separate from capacity; no live utilization |
Service status |
| β οΈ Live unverified | Reader service-policy restrictions; partial data is not proof of global health |
Cluster overview |
| β οΈ Live unverified | Reader physical-capacity policy restrictions; allocations are separate from capacity; no live utilization |
Service log metadata |
| Partial | Service listing only; log retrieval is unavailable without a logging backend. Network agents are listed from the SDK |
Resource metadata |
| Partial | Nova/Neutron/Cinder list/summary only; does not call Gnocchi. Use |
|
| β | Current-project resources; list, attribute filter, show with metrics map, and history passed live |
|
| β | List, per-resource details, show by resource + name, and show by metric UUID passed live |
|
| β | By resource + name and by metric UUID passed live, including start, granularity, resample, refresh, latest-N limit, and |
|
| β | Current-project search is always applied. Passed live: project-wide |
|
| β | All five sections returned live with the admin test account; a denied section is reported per section |
Alarm management |
| β | list/create/show/delete passed live, by name and by ID. Evaluation passed live: |
Compute agents |
| β | list/show of Nova services and hypervisors. |
Usage statistics |
| β | Period compute allocation-hours plus separately labeled current snapshots |
10. π Usage & Quota
OpenStack CLI Command | MCP Tool | Status | Notes |
|
| β | Quota query. The network floating IP quota reads the SDK |
|
| β | Routes supplied fields to Nova/Neutron/Cinder; partial failures are reported |
|
| β | Resets all three service quotas to defaults; requires modification tools |
|
| β | Current-project Nova simple tenant usage; not monetary charges |
|
| β | Limits query |
Resource allocations |
| β οΈ Live unverified | Reader physical-capacity policy restrictions; allocations are separate from capacity; no live utilization |
11. ποΈ Cross-project reads (advanced, opt-in)
Off by default and not needed for single-project use. These rows apply only
when the server sets ALLOW_CROSS_PROJECT_READ=true and OpenStack policy lets
the credential read across projects; see the
cross-project reads guide.
OpenStack CLI Command | MCP Tool | Status | Notes |
|
| β | Requires |
|
| β | Requires |
|
| β | Requires |
|
| β | Requires |
|
| β | Requires |
|
| β | Requires |
|
| β | Requires |
|
| β | Requires |
|
| β | Requires |
|
| β | Requires |
|
| β | Requires |
|
| β | Requires |
|
| β | Requires |
API prerequisites and result semantics
The feature table marks implemented API paths, not verification against every OpenStack deployment. Service and policy requirements must be checked in the target deployment. A missing service, extension, required microversion, or permission is an operation failure, not an empty successful result.
Modification tools: the MCP server registers
set_*management tools only whenALLOW_MODIFY_OPERATIONS=true. Existing action guards also reject writes when disabled. Dedicated read-onlyget_*tools remain available. API authorization is enforced by the target cloud; agent, user, domain, type and QoS administration may require elevated OpenStack roles.Alarms:
set_alarmscalls the Keystone catalog'salarmingendpoint for the configured interface/region. It supports list/create/show/delete of project alarms and createsgnocchi_resources_thresholdalarms. Aodh, Gnocchi, the specified resource and a collected metric are required. The default metriccpu_utilis kept for compatibility, but Ceilometer has not produced it since Stein, so such alarms stayinsufficient data. For CPU alarms, usemetric="cpu"withaggregation_method="rate:mean"; the threshold is CPU nanoseconds per granularity window (percent / 100 Γ granularity Γ 10βΉ Γ vCPUs; 80% of one vCPU over 300 s is2.4e11).metric,resource_type,aggregation_method,granularity,evaluation_periods, andalarm_actionscan be supplied. An empty action list creates an alarm without notifications. There is no simulation fallback.Usage:
get_usage_statisticscalls Nova simple tenant usage for the current project. Requires Nova API microversion 2.40 or newer for complete pagination. Date bounds are UTC midnight, start inclusive and end exclusive; start must precede end.usage_statistics.period_computecontains paginated Nova allocation-hours, not actual CPU load, billed currency, or network/storage history. Nova accounts using VM lifetime and the latest flavor; stop/pause/resize history is not reconstructed. Existingcompute,network, andstoragefields are current snapshots, labeled bysnapshot_fieldsandsnapshot_timestamp. An unavailable period API returns failure. A failed snapshot section is reported as partial failure.Capacity:
get_resource_monitoringseparates flavor allocations from project quota limits and the physical totals of visible hypervisors. Missing physical totals or incomplete flavor allocations returnnull, not invented capacity or zero usage. Physical totals and project allocations have different scopes. This tool does not collect live utilization or trends.Floating IP pools:
total_ipsis the exact sum of visible subnet allocation ranges;used_ipscounts visible Floating IP objects. Globalavailable_ipsisnull: other projects and non-floating ports may consume addresses that are not visible to the caller.Volume groups: the cloud must support Cinder generic groups (microversion 3.13 or newer). Create requires an existing
group_typeand comma-delimitedvolume_types. Deletion does not delete member volumes unlessdelete_volumes=trueis explicitly supplied. Async creation/deletion responses mean accepted, not finished.Identity and images:
set_usersreads a new user's password from the server variable named bypassword_env; never put passwords in prompts. It does not grant project roles. Disable a domain explicitly withset_domains(action="update", enabled=false)before deleting it. Image member modifications require an image owned by the current project, shared visibility, and appropriate Glance policy; adding a member does not automatically accept the recipient's membership.Telemetry (Gnocchi): the read-only
get_gnocchi_resources,get_gnocchi_metrics,get_gnocchi_measures,get_gnocchi_aggregates, andget_gnocchi_service_infotools usegnocchiclientwith the Keystone catalog'smetricendpoint for the configured interface/region. Ceilometer must be collecting data into Gnocchi; it has no API of its own. Results are limited to resources whoseproject_idis the current project. Metrics are created by the Ceilometer service project, so ownership is checked through the embedded resource. Available metrics depend on deployment. Notification-based metrics such asvcpus,memory, anddisk.root.sizerecord flavor sizes. Polled metrics such ascpu,memory.usage,disk.device.*, andnetwork.*require a working ceilometer-compute agent and a running instance.cpuis cumulative nanoseconds; useaggregation="rate:mean"when the archive policy supports it. Current Ceilometer does not producecpu_util. Measures and aggregation methods are limited by each metric's archive policy.get_gnocchi_aggregatesaccepts either structured parameters or an operations expression; both add a current-project resource search, so expression metric references resolve as metric names on matched resources. Arithmetic between metrics returns points only where every operand has data. Gnocchi write operations (measure ingestion, resource/metric/archive policy changes) are not implemented.Barbican and TLS listeners: the Barbican tools use the Keystone catalog's
key-managerendpoint; without it they report "Barbican (key-manager) service is not available". Creating secrets requires the role Barbican's policy demands (adminorcreatorunder the legacy policy, projectmemberunder Secure RBAC). ForTERMINATED_HTTPSlisteners, Octavia's defaultbarbican_acl_authgrants its service user read access to the referenced secret when the listener is created; the requesting user must be allowed to change that secret's ACL. These are deployment prerequisites, described in the Barbican and TLS guide.Logs and telemetry:
set_service_logs(action="show")andset_metrics(action="show")return explicit unsupported errors. No centralized logging backend is implemented.set_metricslist/summary operations expose Nova/Neutron/Cinder resource metadata only; use the Gnocchi tools for telemetry samples.
Contributing
Contributions are welcome. See the development guide for running the tests, adding tools, and the pull request workflow.
License
This project is licensed under the MIT License - see the LICENSE file for details.
Available Tools
61 toolsget_availability_zonesGet Availability ZonesB
List availability zones and their status
Inputs: no named parameters. Returns: result
get_domainsGet DomainsB
List Keystone domains subject to cloud policy.
Inputs: no named parameters. Returns: result
get_floating_ip_poolsGet Floating Ip PoolsA
Get list of floating IP pools (external networks).
Functions:
List all external networks that can provide floating IPs
Show available and used IP counts for each pool
Display network configuration for floating IP allocation
Provide pool capacity and utilization information
Use when user requests:
"Show floating IP pools"
"List available floating IP networks"
"Check floating IP capacity"
"What external networks are available?"
Inputs: no named parameters. Returns: result
get_floating_ipsGet Floating IpsB
Get list of floating IPs with their associations.
Functions:
Query floating IPs and their current status
Display associated fixed IPs and ports
Show floating IP pool and router associations
Provide floating IP allocation and usage information
Use when user requests floating IP information, external connectivity queries, or IP management tasks.
Inputs: all_projects. Returns: result
get_gnocchi_aggregatesGet Gnocchi AggregatesA
Aggregate measured telemetry across many current-project resources.
Use for cross-resource questions: project-wide averages or totals, top N resources, per-flavor or per-host groups, and ratios between metrics. For one resource's time series, use get_gnocchi_measures instead. For allocations, capacity, or quotas, use get_resource_monitoring.
Choose ONE mode:
Simple mode (preferred): set metric and aggregation, plus optional reaggregation, groupby, and top. Leave operations empty.
Project-wide average CPU: metric="cpu", aggregation="rate:mean", reaggregation="mean"
Top 5 instances by CPU: metric="cpu", aggregation="rate:mean", top=5
Mean memory use per flavor: metric="memory.usage", reaggregation="mean", groupby="flavor_name"
Expression mode: set operations only, for arithmetic between metrics or advanced Gnocchi functions. Leave metric and reaggregation empty. Write metric NAMES, never metric UUIDs, as (metric NAME AGGREGATION). Example ratio: (/ (metric memory.usage mean) (metric memory mean)) Arithmetic returns points only where every operand has data; memory (flavor size) is recorded sparsely, so this ratio has fewer points.
Units: cpu with rate:mean is CPU nanoseconds per granularity window; utilization fraction = value / (granularity * 1e9 * vCPUs). memory.usage is MB in use. vcpus and memory are flavor sizes, not usage.
Results are always limited to resources of the current project. The response echoes the executed operations expression.
Inputs: top, fill, stop, limit, +10 more. Returns: result
get_gnocchi_measuresGet Gnocchi MeasuresA
Read stored Gnocchi measures (telemetry time series) for a metric.
Use for measured utilization, load, trends, and history ("CPU usage of vm1", "memory over the last hour"). Values are observed telemetry collected by Ceilometer, such as cpu, memory.usage, or disk and network counters. Cumulative counters like cpu (nanoseconds) grow over time; compare consecutive values for rates. Some metrics such as vcpus and memory record flavor sizes, not usage. For averages, totals, top N, or groups across many resources, use get_gnocchi_aggregates. For allocations, capacity, or quotas, use get_resource_monitoring or get_quota. If a metric is not collected, report that; do not substitute allocation data.
Inputs: metric (required), stop, limit, start, +5 more. Returns: result
get_gnocchi_metricsGet Gnocchi MetricsA
List or show current-project Gnocchi metrics.
Without arguments, lists metric names and IDs of current-project resources. With resource_id, shows that resource's metrics including archive policy and unit. With metric, shows one metric. Use it to check which telemetry metrics are actually collected before reading measures or creating alarms. These are measured telemetry series, not allocations or quotas.
Inputs: limit, metric, resource_id, resource_type. Returns: result
get_gnocchi_resourcesGet Gnocchi ResourcesA
List, search, or show current-project Gnocchi (telemetry) resources.
Resources are the objects Ceilometer measures, such as instances, disks, and network interfaces. Each resource maps metric names to metric IDs. Use this to find which metrics exist for a server before reading measures.
Inputs: limit, marker, filters, history, +2 more. Returns: result
get_gnocchi_service_infoGet Gnocchi Service InfoA
Read Gnocchi service configuration and health information.
Use to check which granularities and aggregation methods a metric supports (for get_gnocchi_measures or set_alarms), which resource types exist, and whether measure processing is backlogged.
Inputs: sections. Returns: result
get_heat_stack_detailsGet Heat Stack DetailsB
Query resources, events, template, or outputs for a Heat stack.
Inputs: stack_name (required), section, nested_depth. Returns: result
get_heat_stacksGet Heat StacksA
Get list of Heat orchestration stacks.
Functions:
Query Heat stacks and their current status
Display stack creation and update timestamps
Use get_heat_stack_details for templates, resources, events, or outputs
Provide orchestration deployment information
Use when user requests stack information, orchestration queries, or infrastructure-as-code status.
Inputs: all_projects. Returns: result
get_hypervisor_detailsGet Hypervisor DetailsC
Get detailed information about hypervisors
Inputs: hypervisor_name. Returns: result
get_identity_groupsGet Identity GroupsC
List Keystone groups subject to cloud policy.
Inputs: no named parameters. Returns: result
get_image_detail_listGet Image Detail ListB
Get detailed list of all images with comprehensive metadata.
Functions:
List all images available in the project
Show image status, size, and format information
Display image properties and metadata
Provide ownership and visibility details
Use when user requests image listing, image information, or image metadata details.
Inputs: all_projects. Returns: result
get_image_membersGet Image MembersC
List member projects for an image.
Inputs: image_name (required). Returns: result
get_instanceGet InstanceA
Unified instance query tool supporting all instance retrieval patterns. Consolidates functionality from get_instance_details, get_instance_by_name, get_instances_by_status, and search_instances.
Functions:
Get specific instances by names or IDs
Filter instances by status (ACTIVE, SHUTOFF, ERROR, etc.)
Search instances across multiple fields (name, flavor, image, host, etc.)
List all instances with pagination
Support both summary and detailed information modes
Use when user requests instance information, status checks, or instance searches.
Inputs: ids, limit, names, offset, +6 more. Returns: result
get_instance_by_nameGet Instance By NameB
Get detailed information for a specific instance by name.
Inputs: instance_name (required). Returns: result
get_instance_detailsGet Instance DetailsB
Provides detailed information and status for OpenStack instances with pagination support.
Functions:
Query basic instance information (name, ID, status, image, flavor) with efficient pagination
Collect network connection status and IP address information
Check CPU, memory, storage resource usage and allocation
Provide instance metadata, keypair, and security group settings
Support large-scale environments with configurable limits
Use when user requests specific instance information, VM details, server analysis, or instance troubleshooting.
Inputs: limit, offset, include_all, all_projects, +3 more. Returns: result
get_instances_by_statusGet Instances By StatusC
Get instances filtered by status.
Inputs: status (required), all_projects. Returns: result
get_key_manager_quotasGet Key Manager QuotasA
Show the effective Barbican quotas (secrets, orders, containers, consumers, CAs) of the current project; -1 means unlimited.
Inputs: no named parameters. Returns: result
get_keypair_listGet Keypair ListA
Get list of SSH keypairs for the current user.
Functions:
Query SSH keypairs and their fingerprints
Display keypair types and creation dates
Show public key information (truncated for security)
Provide keypair management information
Use when user requests SSH key management, keypair information, or security key queries.
Inputs: no named parameters. Returns: result
get_load_balancer_amphoraeGet Load Balancer AmphoraeA
Get amphora instances for a load balancer or all amphorae.
Inputs: lb_name_or_id. Returns: result
get_load_balancer_availability_zonesGet Load Balancer Availability ZonesC
Get load balancer availability zones.
Inputs: no named parameters. Returns: result
get_load_balancer_detailsGet Load Balancer DetailsA
Get detailed information about a specific OpenStack load balancer.
Functions:
Shows comprehensive load balancer details including VIP configuration
Lists all listeners with their protocols and ports
Shows pools and members for each listener
Displays health monitor information if configured
Provides provisioning and operating status
Use when user requests:
"Show details for load balancer [name/id]"
"Get load balancer configuration"
"Show load balancer listeners and pools"
"What's the status of load balancer [name]?"
Inputs: lb_name_or_id (required). Returns: result
get_load_balancer_flavorsGet Load Balancer FlavorsC
Get load balancer flavors.
Inputs: no named parameters. Returns: result
get_load_balancer_health_monitorsGet Load Balancer Health MonitorsA
Get health monitors, optionally filtered by pool.
Functions:
Lists all health monitors or monitors for a specific pool
Shows monitor types (HTTP, HTTPS, TCP, PING, UDP-CONNECT)
Displays health check intervals, timeouts, and retry settings
Provides HTTP-specific settings (method, URL path, expected codes)
Use when user requests:
"Show all health monitors"
"List health monitors for pool [name/id]"
"What health checks are configured?"
"Show health monitor configuration"
Inputs: pool_name_or_id. Returns: result
get_load_balancer_l7_policiesGet Load Balancer L7 PoliciesA
Get L7 policies for a listener or all L7 policies.
Inputs: listener_name_or_id. Returns: result
get_load_balancer_l7_rulesGet Load Balancer L7 RulesC
Get L7 rules for a specific L7 policy.
Inputs: policy_name_or_id (required). Returns: result
get_load_balancer_listGet Load Balancer ListA
Retrieve comprehensive list of OpenStack load balancers with detailed information.
Functions:
Lists all load balancers in the OpenStack cluster
Provides detailed load balancer information including VIP, status, listeners
Supports pagination for large environments (limit/offset)
Shows listener count and basic listener information for each load balancer
Displays provisioning and operating status for troubleshooting
Use when user requests:
"Show me all load balancers"
"List load balancers with details"
"What load balancers are available?"
"Show load balancer status"
Inputs: limit, offset, include_all, all_projects. Returns: result
get_load_balancer_listenersGet Load Balancer ListenersA
Get listeners for a specific OpenStack load balancer.
Functions:
Lists all listeners attached to a load balancer
Shows listener protocols, ports, and configurations
Displays admin state and default pool associations
Provides creation and update timestamps
Use when user requests:
"Show listeners for load balancer [name/id]"
"List load balancer listeners"
"What ports are configured on load balancer [name]?"
"Show listener configuration for [lb_name]"
Inputs: lb_name_or_id (required). Returns: result
get_load_balancer_pool_membersGet Load Balancer Pool MembersA
Get members for a specific OpenStack load balancer pool.
Functions:
Lists all members in a specific pool
Shows member addresses, ports, weights, and health status
Displays member admin state and operational status
Provides monitor configuration for each member
Use when user requests:
"Show members for pool [name/id]"
"List pool members"
"What members are in pool [name]?"
"Show pool member status"
Inputs: pool_name_or_id (required). Returns: result
get_load_balancer_poolsGet Load Balancer PoolsA
Get load balancer pools, optionally filtered by listener.
Functions:
Lists all pools or pools for a specific listener
Shows pool protocols, load balancing algorithms
Displays members in each pool with their status
Provides health monitor associations
Use when user requests:
"Show all load balancer pools"
"List pools for listener [name/id]"
"What pools are configured on [listener_name]?"
"Show pool members and their status"
Inputs: listener_name_or_id. Returns: result
get_load_balancer_providersGet Load Balancer ProvidersC
Get load balancer providers.
Inputs: no named parameters. Returns: result
get_load_balancer_quotasGet Load Balancer QuotasA
Get load balancer quotas for a project or all projects.
Inputs: project_id. Returns: result
get_network_detailsGet Network DetailsA
Provides detailed information for OpenStack networks, subnets, routers, and security groups.
Functions:
Query configuration information for specified network or all networks
Check subnet configuration and IP allocation status per network
Collect router connection status and gateway configuration
Analyze security group rules and port information
Use when user requests network information, subnet details, router configuration, or network troubleshooting.
Inputs: all_projects, network_name. Returns: result
get_network_qos_policiesGet Network Qos PoliciesA
List project-owned Neutron QoS policy containers; no bandwidth rules are created.
Inputs: no named parameters. Returns: result
get_project_detailsGet Project DetailsC
Get OpenStack project details (similar to 'openstack project list/show').
Inputs: all_projects, project_name. Returns: result
get_quotaGet QuotaA
Get quota information for projects (similar to 'openstack quota show').
Each service reports limits, usage and usage_percent (percent of each limit in use). A usage value is null when it could not be determined, and usage_percent is null for unlimited (-1), zero or missing limits and for unknown usage.
Inputs: project_name. Returns: result
get_resource_monitoringGet Resource MonitoringA
Read current project allocations and visible hypervisor capacity.
Functions:
Report project flavor allocations, not live CPU or memory utilization
Collect hypervisor statistics and resource allocation
Separate project quota limits from visible physical capacity
Return null for unavailable capacity or incomplete flavor allocations
Use for current inventory and capacity planning. No historical trends or telemetry samples are collected. Physical totals and project allocations have different scopes and should not be divided into a utilization rate. For measured CPU/memory utilization, load, or trends, use get_gnocchi_measures instead; never present these allocations as usage.
Inputs: no named parameters. Returns: result
get_role_assignmentsGet Role AssignmentsA
Get role assignments for the current project.
Functions:
Query role assignments for users and groups
Display project-level and domain-level permissions
Show scope of role assignments
Provide comprehensive access control information
Use when user requests permission information, access control queries, or security auditing.
Inputs: no named parameters. Returns: result
get_routersGet RoutersB
Get list of routers with their configuration.
Functions:
Query routers and their external gateway configurations
Display router interfaces and connected networks
Show routing table entries and static routes
Provide comprehensive network routing information
Use when user requests router information, network connectivity queries, or routing configuration.
Inputs: all_projects. Returns: result
get_secret_aclGet Secret AclA
Show the read ACL of a Barbican secret or container.
The ACL lists users outside the owning project allowed to read it and whether project-wide access (project-access) is enabled.
Inputs: name_or_ref (required), target_type. Returns: result
get_secret_containersGet Secret ContainersB
List or show Barbican secret containers of the current project.
A container groups secret hrefs, e.g. a certificate container with certificate, private_key and intermediates entries. Payloads are never returned.
Inputs: name, limit. Returns: result
get_secret_ordersGet Secret OrdersA
List or show Barbican key-generation orders of the current project.
An ACTIVE order points to the generated secret (key orders) or container (asymmetric orders) through secret_ref or container_ref.
Inputs: limit, order_id. Returns: result
get_secretsGet SecretsA
List or show Barbican secrets of the current project (metadata only).
Never returns a secret payload; payload retrieval is intentionally not provided so key material cannot reach the LLM. Use the secret_ref from this tool as a reference, for example in set_load_balancer_listener.
Inputs: name, limit, secret_type. Returns: result
get_secret_storesGet Secret StoresA
List Barbican secret store backends. Only clouds that enable multiple secret store backends serve this API; others return an error.
Inputs: no named parameters. Returns: result
get_security_groupsGet Security GroupsA
Get list of security groups with their rules.
Functions:
Query security groups and their rule configurations
Display ingress and egress rules with protocols and ports
Show remote IP prefixes and security group references
Provide comprehensive network security information
Use when user requests security group information, firewall rules, or network security queries.
Inputs: all_projects. Returns: result
get_server_eventsGet Server EventsA
Get recent events for a server by name or UUID.
Pass either value using instance_name, for example get_server_events(instance_name=""). There is no server_id parameter; do not use server_id or instance_id as argument names.
Inputs: instance_name (required), limit, include_event_details. Returns: result
get_server_groupsGet Server GroupsC
List all server groups with their details
Inputs: all_projects. Returns: result
get_server_volumesGet Server VolumesC
Get all volumes attached to a specific server
Inputs: instance_name (required). Returns: result
get_service_statusGet Service StatusA
Provides status and health check information for each OpenStack service.
Functions:
Check active status of all OpenStack services
Verify API endpoint responsiveness for each service
Collect detailed status and version information per service
Detect and report service failures or error conditions
Use when user requests service status, API status, health checks, or service troubleshooting.
Inputs: no named parameters. Returns: result
get_usage_statisticsGet Usage StatisticsA
Get current-project Nova allocation-hours and separate current snapshots.
Functions:
Show project usage statistics over a specified time period
Display servers, RAM MB-Hours, CPU Hours, and Disk GB-Hours
Provide detailed server usage breakdown when available
Keep current compute/network/storage snapshots separate from period data
Use for project usage reports. Period data comes from Nova simple tenant usage. It measures allocation-hours, not live utilization or monetary charges. For measured CPU/memory utilization or trends, use get_gnocchi_measures. Missing or denied usage API returns failure without synthetic fallback.
Inputs: end_date, start_date. Returns: result
get_user_listGet User ListA
Get list of OpenStack users in the current domain.
Functions:
Query user accounts and their basic information
Display user status (enabled/disabled)
Show user email and domain information
Provide user creation and modification timestamps
Use when user requests user management information, identity queries, or user administration tasks.
Inputs: no named parameters. Returns: result
get_volume_backupsGet Volume BackupsB
List Cinder backups visible in the current project.
Inputs: all_projects. Returns: result
get_volume_groupsGet Volume GroupsA
List current-project Cinder generic groups. Requires API microversion >= 3.13.
Inputs: all_projects. Returns: result
get_volume_listGet Volume ListA
Get list of all volumes with detailed information.
Functions:
List all volumes in the project
Show volume status, size, and type information
Display attachment information for volumes
Provide detailed metadata for each volume
Use when user requests volume listing, volume information, or storage overview.
Inputs: all_projects. Returns: result
get_volume_qosGet Volume QosA
List Cinder QoS specifications, or show one specification's volume type associations by name or ID. Cloud policy may require admin rights.
Inputs: qos_name. Returns: result
get_volume_snapshotsGet Volume SnapshotsA
Get list of volume snapshots.
Functions:
Query volume snapshots and their status
Display source volume information
Show snapshot creation and modification dates
Provide snapshot size and usage information
Use when user requests snapshot information, backup queries, or volume restoration planning.
Inputs: all_projects. Returns: result
get_volume_transfersGet Volume TransfersA
List current-project Cinder volume transfer requests without authorization keys.
Inputs: no named parameters. Returns: result
get_volume_typesGet Volume TypesA
Get list of volume types with their specifications.
Functions:
Query volume types and their capabilities
Display extra specifications and backend configurations
Show public/private volume type settings
Provide storage backend information
Use when user requests volume type information, storage backend queries, or volume creation planning.
Inputs: no named parameters. Returns: result
search_instancesSearch InstancesB
Search for OpenStack instances based on various criteria with efficient pagination.
Functions:
Search instances by name, status, host, flavor, image, or availability zone
Support partial matching with configurable case sensitivity
Return detailed information for matching instances with pagination
Optimized for large-scale environments with intelligent filtering
Inputs: search_term (required), limit, offset, search_in, +2 more. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelogβ¦
TDQS
Scored across 61 tools
The instance tools overlap heavily: get_instance is explicitly a unified replacement for get_instance_details, get_instance_by_name, get_instances_by_status, and search_instances, yet all of those legacy tools remain exposed. Other areas are better separated, but the retained redundant instance tools create multiple unclear boundaries for an agent.
Almost all tools use a predictable get_snake_case pattern, with consistent resource grouping such as get_load_balancer_* and get_volume_*. The main deviation is search_instances, and some names awkwardly append list/detail/list, but overall naming is readable and mostly consistent.
With 61 tools, the surface is far beyond the typical well-scoped range and well into the 50+ extreme-mismatch category. Granular read-only queries for load balancer subresources, Gnocchi telemetry, Barbican secrets, and many OpenStack services make the toolset unwieldy for an agent.
The tool surface is almost entirely read-only get/search operations, with no create, update, delete, or lifecycle action tools for instances, networks, volumes, load balancers, Heat stacks, secrets, keypairs, or users. For an OpenStack ops server, this is a severe gap that prevents most operational workflows.
Maintenance
Related MCP Connectors
One MCP server for all your MCP servers: code mode, per-tool permissions, approvals and audit.
A MCP server built for developers enabling Git based project management with project and personalβ¦
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceAn MCP server that enables interaction with multiple Kubernetes clusters simultaneously, providing comprehensive tools for cluster management, resource operations, and diagnostics across different environments.-
- FlicenseCqualityDmaintenanceMCP server that enables deployment and management of MCP servers through a simple configuration-based interface.1-
- AlicenseBqualityDmaintenanceAn MCP server for Proxmox VE that enables read-only cluster inspection, VM/container lifecycle operations, snapshots, migration, and provisioning with safe confirmation gates.28MIT
- AlicenseCqualityFmaintenanceMCP server for project management automation, providing tools for project health, documentation, task management, security scanning, and CI/CD validation.271MIT