{ "SchemaVersion": "1", "Kind": "DirectoryIssues", "Slug": "io-github-victoriametrics-community-mcp-victoriametrics", "Name": "io.github.VictoriaMetrics-Community/mcp-victoriametrics", "CanonicalUrl": "https://askpod.ai/mcp/io-github-victoriametrics-community-mcp-victoriametrics/issues", "ServerUrl": "https://askpod.ai/mcp/io-github-victoriametrics-community-mcp-victoriametrics", "IssueTotal": 19, "Held": 17, "Issues": [ { "Title": "The `documentation` tool is too attractive for agents", "Excerpt": "Whenever I have the MCP server enabled and ask agents questions completely unrelated to VictoriaMetrics, they try to use the `documentation` tool. For example here I asked Opus 4.8 a question about Backrest:\n\n\"Image\"\n\nWhat do you think about renaming this tool to contain VictoriaMetrics in its name?", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/186", "PublishedAt": "2026-06-16T04:07:45.000Z", "State": "open", "Comments": 0, "Reporter": "External", "Rank": "recent", "Extractor": "github_issue" }, { "Title": "Alerts tool panics when limit is greater than returned alerts count", "Excerpt": "The `alerts` tool panics when it is called with a `limit` value greater than the number of alerts returned by the VictoriaMetrics API.\n\nRequest:\n\n```json\n{\n \"state\": \"firing\",\n \"limit\": 50\n}\n```\n\nIn my MCP client this tool is exposed as:\n\n```text\nmcp_mcp_victoriametrics_alerts\n```\n\nError:\n\n```text\n2026-05-19 13:53:02,818 ERROR [20260519_135155_2a8e0a] tools.mcp_tool: MCP tool mcp-victoriametrics/alerts call failed: panic recovered in alerts tool handler: runtime error: slice bounds out of rang", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/175", "PublishedAt": "2026-05-19T15:38:54.000Z", "State": "open", "Comments": 0, "Reporter": "External", "Rank": "recent", "Extractor": "github_issue" }, { "Title": "Web UI is not available on docker install", "Excerpt": "Running MCP in docker doesn't render a Web UI:\n```\n│ vmm-victoria-metrics-mcp-66bb85f877-6zzg2 time=2026-05-15T13:10:47.106Z level=INFO msg=\"Starting mcp-victoriametrics\" version=1.19.0 date=2026-03-17T12:55:19Z mode=http addr=0.0.0.0:8080 │\n│ vmm-victoria-metrics-mcp-66bb85f877-6zzg2 time=2026-05-15T13:10:53.702Z level=INFO msg=\"Starting server in HTTP mode\" addr=0.0.0.0:8080 │\n│ vmm-victoria-metrics-mcp-66bb85f877-6zzg2 time=2026-05-15T13:10:5", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/172", "PublishedAt": "2026-05-15T13:19:04.000Z", "State": "closed", "Comments": 3, "Reporter": "External", "Rank": "top", "Extractor": "github_issue" }, { "Title": "Support dynamically selecting a host per tool call", "Excerpt": "We have multiple victoria metrics hosts, it would be nice if only a single MCP is needed in order to send tool calls off to a different host. Could be done as optional parameters on the tool call itself, or since in our case we use streamable-http a header would suffice", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/136", "PublishedAt": "2026-03-12T17:27:43.000Z", "State": "open", "Comments": 3, "Reporter": "External", "Rank": "top", "Extractor": "github_issue" }, { "Title": "can not send query request to vmgateway if it enable auth", "Excerpt": "Reproduce\nHaving VM cluster with vmgateway enable JWT token auth. Within this, the request don't need /prometheus prefix to be pass to vmgateway. However the mcp still sending request with prefix. So vmgateway repsone with 404\nI tried to make a patch for fix this by removing the prefix, then the MCP works", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/130", "PublishedAt": "2026-03-07T13:20:38.000Z", "State": "open", "Comments": 0, "Reporter": "External", "Rank": "recent", "Extractor": "github_issue" }, { "Title": "Support dynamic per-request headers in tool calls", "Excerpt": "## Problem\n\n`VM_INSTANCE_HEADERS` is static — set once at startup and cannot be changed during the MCP server's lifetime.\n\nThis doesn't work when headers need to be dynamic, for example short-lived auth tokens that expire and must be refreshed.\n\n## Typical Scenario\n\nVictoriaMetrics cluster deployed behind an auth proxy:\n\n```\nMCP Client → mcp-victoriametrics → Auth Proxy (validates X-Token) → vmselect\n```\n\n- The auth proxy requires a custom `X-Token` header on every request\n- Tokens are short-liv", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/129", "PublishedAt": "2026-03-04T09:35:13.000Z", "State": "closed", "Comments": 13, "Reporter": "External", "Rank": "top", "Extractor": "github_issue" }, { "Title": "The default tenant parameter for queries is 0, modify it to set the tenant parameter via environment variable", "Excerpt": "The default tenant parameter for queries in cluster mode is hardcoded to '0', which causes queries to fail when the actual tenant ID is different. Need to modify the code to allow setting the default tenant via an environment variable (e.g., VM_DEFAULT_TENANT) to support different tenant configurations.\n\nhttps://xxxx.com/select/0/prometheus/api/v1/", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/113", "PublishedAt": "2026-01-06T10:09:43.000Z", "State": "closed", "Comments": 2, "Reporter": "External", "Rank": "recent", "Extractor": "github_issue" }, { "Title": "Installation Error Using Smithery", "Excerpt": "I'm receiving the following error when trying to install the mcp server using smithery.\n\n`npx -y @smithery/cli install @VictoriaMetrics-Community/mcp-victoriametrics --client vscode`\n\n✖ Failed to install @VictoriaMetrics-Community/mcp-victoriametrics\nError: API error occurred: {\"error\":\"Server not found\",\"request$\":{},\"response$\":{},\"body$\":\"{\\\"error\\\":\\\"Server not found\\\"}\"}", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/109", "PublishedAt": "2025-12-18T19:57:20.000Z", "State": "closed", "Comments": 1, "Reporter": "External", "Rank": "recent", "Extractor": "github_issue" }, { "Title": "Can We Have Accessible Helm Chart (e.g. github pages)", "Excerpt": "I saw the helm chart was there in `k8s/helm` but I wonder, maybe we can have it accessible through Github Pages?", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/102", "PublishedAt": "2025-12-05T02:43:52.000Z", "State": "closed", "Comments": 3, "Reporter": "External", "Rank": "top", "Extractor": "github_issue" }, { "Title": "Feature request: add otel-collector receivers metrics to metrics_metadata_db", "Excerpt": "Hi VM team,\n\nwe're using otel-collector to collect metrics from node, kubelet and k8s apiserver with following receivers:\n- [hostmetricsreceiver](https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/main/receiver/hostmetricsreceiver)\n- [kubeletstatsreceiver](https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/main/receiver/kubeletstatsreceiver)\n- [k8sclusterreceiver](https://github.com/open-telemetry/opentelemetry-collector-contrib/tree/main/receiver/k8sclust", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/90", "PublishedAt": "2025-11-07T07:18:43.000Z", "State": "closed", "Comments": 10, "Reporter": "External", "Rank": "top", "Extractor": "github_issue" }, { "Title": "Error adding server to MCP Context Forge", "Excerpt": "Hi team,\n\nI experience issues trying to add this MCP server to my IBM Context Forge app. I've deployed my Context Forge app and this server in the same Kubernetes cluster and the same namespace. The versions are the latest for each product.\n\nI launch the server with the following config : \n\n- VM_INSTANCE_ENTRYPOINT = http://\\\n- VM_INSTANCE_TYPE = cluster\n- MCP_SERVER_MODE = http\n- MCP_LISTEN_ADDR = 0.0.0.0:8080\n\nI should be able to add this server by targeting the url **http://", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/78", "PublishedAt": "2025-10-22T08:51:03.000Z", "State": "closed", "Comments": 2, "Reporter": "External", "Rank": "recent", "Extractor": "github_issue" }, { "Title": "Hi team I get a strange regex error when using Gemini CLI as client", "Excerpt": "\"Image\"\n\nHi team I get the error as above when I use Gemini CLI.\nThis error seems to be not happen when I attach it to Github Copilot though", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/74", "PublishedAt": "2025-10-13T16:26:36.000Z", "State": "closed", "Comments": 2, "Reporter": "External", "Rank": "recent", "Extractor": "github_issue" }, { "Title": "Running container as non-root user", "Excerpt": "**Summary**\nThe current Docker image runs as root. This is not a best practice to run container in Kubernetes. Please update the Dockerfile to add and run as a dedicated non-root user instead of root.\n\n**Risks of running as root**\n\nPrivilege escalation if the container is broken out of.\n\nFile permission surprises when shared volumes are used.\n\nIncompatible with restrictive clusters (PSa/PSP, OPA/Gatekeeper, PodSecurity “restricted” profile).\n\n**Proposal**\n\nUpdate the Dockerfile to create a non-r", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/58", "PublishedAt": "2025-09-17T05:08:10.000Z", "State": "closed", "Comments": 3, "Reporter": "External", "Rank": "recent", "Extractor": "github_issue" }, { "Title": "Error running in streamable http mode", "Excerpt": "Hi Team,\n\nWhile running the server in streamable http mode, getting below error:\n\n**Error on server:**\n`2025/09/15 05:53:18 Starting mcp-victoriametrics version dev (date: unknown)\n2025/09/15 05:53:26 Starting server in HTTP mode on 0.0.0.0:8080\n2025/09/15 05:53:26 Server is listening on 0.0.0.0:8080\n2025/09/15 05:54:03 ERROR: Failed to deliver sampling response: no pending request found for session mcp-session-2ad06bc0-2dcd-408e-8c13-e9e07bf2d4d7, request 1\n2025/09/15 05:54:03 ERROR: Failed to ", "SourceUrl": "https://github.com/VictoriaMetrics/mcp-victoriametrics/issues/51", "PublishedAt": "2025-09-15T06:00:19.000Z", "State": "closed", "Comments": 9, "Reporter": "External", "Rank": "top", "Extractor": "github_issue" }, { "Title": "Alert tool does not work with local models", "Excerpt": "When using the alerts tool with both qwen3:14b and gpt-oss:20b and witsy as the client I receive a chain of errors like the following, then the prompt either returns there are no alerts or that it was not able to read the response from the tool call, but using the same prompt with an anthropic model the response is parsed correctly and the output is returned to the user.\n\n