About OneUptime
OneUptime is an MCP server published by OneUptime in the Security category: complete open-source monitoring and observability platform.
The repository has 7.8K stars and 489 forks, with the last commit 6 hours ago. Recent commit activity is the strongest signal that reported issues will get attention.
Install
git clone https://github.com/OneUptime/oneuptimeThis server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.
Tools (54)
acknowledge_alertAcknowledge an alert: marks it as being worked on by moving it to the project
Parameters not published.
acknowledge_incidentAcknowledge an incident: marks it as being worked on by moving it to the project
Parameters not published.
add_alert_affected_resourcesLink an existing alert to the infrastructure and services it is about, as its Affected Resources do: OneUptime AI investigates and fixes what an alert is linked to. What it is linked to already stays. Name each resource in affectedResources by its OneUptime ID from query_telemetry_resources (or lookup_context for a service). alertId defaults to the alert this conversation is about; asked from a re
Parameters not published.
add_alert_noteAdd an internal note to an alert, visible to your team. Markdown is supported.
Parameters not published.
add_incident_affected_resourcesLink an existing incident to the infrastructure and services it is about, as its Affected Resources do: OneUptime AI investigates and fixes what an incident is linked to. What it is linked to already stays. Name each resource in affectedResources by its OneUptime ID from query_telemetry_resources (or lookup_context for a service). incidentId defaults to the incident this conversation is about; ask
Parameters not published.
add_incident_noteAdd a note to an incident. Internal notes are visible to your team only; public notes are posted to the status page for subscribers/customers. Markdown is supported.
Parameters not published.
change_incident_severityChange the severity of an existing incident. Provide the incident id and the new severity name (must match an existing incident severity in this project, e.g.
Parameters not published.
commit_code_to_branchCommit file changes to an EXISTING branch of a connected repository on GitHub, GitLab, Forgejo or Gitea. Never commits to the default or a protected branch — those are refused. Use this only when the user names a branch to commit onto; otherwise prefer open_code_pull_request. Read each file on that branch first, with read_code_file and branchName: content replaces the whole file.
Parameters not published.
create_alertRaise (create) a new alert in this project, as the dashboard
Parameters not published.
create_incidentDeclare (create) a new incident in this project. Provide a clear title and a description of the problem. Optionally set the severity by name (must match an existing incident severity); otherwise the project
Parameters not published.
Show all 54 tools
execute_remediation_commandNo description in manifest
Parameters not published.
find_code_for_exceptionGiven an exception id (from top_exceptions), work out which connected repository its code lives in and which source files and line numbers its stack trace implicates. This is the bridge from a telemetry signal to the code that caused it: call it before read_code_file when investigating an exception
Parameters not published.
find_production_exceptionsExceptions this repository
Parameters not published.
find_production_incidentsIncidents linked to this repository: those that affected one of its services, and those a OneUptime fix in this repository was opened for - newest first, with state, severity, root cause and postmortem. Pass paths (repository files) or searchText to keep only incidents whose title, root cause, remediation or postmortem mention them. Private incidents are never listed.
Parameters not published.
get_ai_investigationGet the autonomous AI investigation (root cause analysis) for one incident or alert. ALWAYS call this FIRST — before re-deriving a root cause from raw telemetry — when the user asks about an incident or alert: an autonomous investigation may already have analyzed it and posted a cited RCA to its timeline. Pass exactly one of incidentId (from query_incidents) or alertId (from query_alerts). Returns
Parameters not published.
get_change_detailsGet one recorded change by its changeId (from recent_changes, or the Recent changes section of an investigation
Parameters not published.
get_data_table_freshnessWhether a table Data Observability watches is fresh, before you use its data: fresh or late by its own learned rhythm, when it last changed and how often it usually does, its open alerts, the pipelines that write it with how their latest run went, and the data products it is part of. Pass table: its schema.table key (public.orders), its name alone (orders), or its ID. A name several databases shar
Parameters not published.
get_public_status_page_announcementsGet announcements from a public status page.
Parameters not published.
get_public_status_page_incidentsGet incidents from a public status page.
Parameters not published.
get_public_status_page_overviewGet the complete overview of a public status page including current status, resources, active incidents, scheduled maintenance, and announcements.
Parameters not published.
get_public_status_page_scheduled_maintenanceGet scheduled maintenance events from a public status page.
Parameters not published.
get_traceThe span tree of one production trace by traceId - each span
Parameters not published.
list_code_repositoriesList the code repositories connected to this project, with the repositoryId needed by search_code and read_code_file. Call this first when the user asks about source code and you do not already know which repository to look in.
Parameters not published.
list_command_targetsNo description in manifest
Parameters not published.
log_histogramGet log volume over time, split by severity. Use this to spot error spikes and quiet periods before drilling into raw logs.
Parameters not published.
lookup_contextDiscover what exists in this project so other tools can be called with exact identifiers: telemetry services (name → ID), metric names, and attribute keys for logs/traces/metrics — for one metric too, to know what query_metrics can filter or group it by. ALWAYS resolve a service name to its ID here before filtering other tools by service.
Parameters not published.
oneuptime_describe_resourceGet the fields of one OneUptime resource: what each means, and how to filter, sort and select on it. Use it before filtering or sorting a list_ or count_ tool, or to understand a field of a create_ or update_ tool.
Parameters not published.
oneuptime_helpGet help and guidance for using the OneUptime MCP server. Returns information about available resources and common operations. Use this tool first to understand what you can do with OneUptime.
Parameters not published.
oneuptime_list_resourcesList all available OneUptime resources and their supported operations. Use this to discover what resources you can manage through the MCP server.
Parameters not published.
oneuptime_whoamiGet the project your credentials belong to (ID and name), and - when signed in with OAuth - who you are signed in as and whether you may make changes. Call this first to understand your context. Note: create tools infer projectId from your credentials automatically, so you never need to pass it.
Parameters not published.
open_code_pull_requestPropose a code change: creates a new branch off the default branch, commits the changes, and opens a pull request (a merge request on GitLab), ready for review, for a human to review and merge. Works in repositories on GitHub, GitLab, Forgejo and Gitea. This is the safe, preferred way to change code from chat — it never writes to the default branch and never merges. Read each file on the default b
Parameters not published.
page_on_call_policyPage (trigger) an on-call duty policy so its responders are notified about an incident. Provide the on-call policy id and the incident id it relates to. Find the policy id with query_on_call_policies and the incident id with query_incidents first. This notifies real people — only do it when clearly asked.
Parameters not published.
post_incident_status_updatePost a customer-facing status update (public note) on an incident. It appears on the status page and, by default, notifies subscribers - unless subscribers were not notified when the incident was declared, in which case it stays quiet by default. A private incident takes no public notes: for one, write a private note with create_incident_note instead. Provide the incident id and the update text.
Parameters not published.
propose_remediation_commandsThe ordered commands to run once approved. Keep the plan minimal.
Parameters not published.
query_ai_insightsList recent AI Insights — preventive findings from OneUptime AI
Parameters not published.
query_alertsQuery alerts in this project with their current state and severity. Every row also carries the monitor the alert was raised for — or, for an alert raised by an SLO burn rate rule, the SLO (slos) — so counting rows by monitor or SLO answers
Parameters not published.
query_cloud_accountsThe cloud accounts connected to OneUptime - AWS accounts, Azure subscriptions, Google Cloud projects - with the name their cloud gives each, how many resources each holds by type and region, and when each was last read and last listed in full. Use it to answer what is in the project
Parameters not published.
query_cloud_resourceWhat a connected cloud (AWS, Azure, Google Cloud) says about one of its resources, from the account
Parameters not published.
query_data_security_issuesData Security issues: what is wrong with how a data store keeps its data - a public bucket, a database open to the internet, no encryption or backups, secrets or plaintext passwords in a store, sensitive data in logs, data outside its allowed regions - each with a severity and a risk score from 0 to 100. Pass dataSecurityIssueId for one issue with its evidence, why it matters and how to fix it on
Parameters not published.
query_data_store_accessWho and what can reach Data Security
Parameters not published.
query_data_storesData Security\
Parameters not published.
query_incidentsQuery incidents in this project. Returns the most recent incidents with their current state and severity. Pass state=
Parameters not published.
query_metricsA metric one of this repository
Parameters not published.
query_tracesLatency and error profile of one of this repository
Parameters not published.
read_code_fileRead the source of one file from a connected repository, with line numbers. Pass startLine/endLine to read the region around a stack-trace line rather than the whole file. Quote only the few lines that matter in your answer. It reads the repository
Parameters not published.
recent_deploysDeploys, rollouts, config changes and rollbacks recorded for this repository
Parameters not published.
resolve_alertResolve an alert: moves it to the project
Parameters not published.
resolve_incidentResolve an incident: moves it to the project
Parameters not published.
run_runbookStart (execute) a runbook by its id, optionally linked to an incident. Use this to run a predefined remediation or diagnostic automation. Use the runbook id the user gives, or find it (and read its steps) with query_runbooks first.
Parameters not published.
search_codeFind source file paths in a connected repository by matching part of a file or directory name (e.g.
Parameters not published.
search_incidentsSearch PAST incidents by free text across their title, description, root cause, remediation notes and postmortem. Use this to ground root-cause analysis in history — e.g.
Parameters not published.
search_logsProduction log lines of one of this repository
Parameters not published.
start_investigationStart (enqueue) an autonomous AI investigation for one incident or alert. Use this when get_ai_investigation shows no investigation exists — the automatic gates (severity floor, dedupe window) may have skipped the subject — and the user wants an AI root cause analysis. Pass exactly one of incidentId (from query_incidents) or alertId (from query_alerts). The run is queued durably, usually completes
Parameters not published.
top_exceptionsGet the top exceptions (grouped by fingerprint) in this project, sorted by occurrence count. Includes first/last seen timestamps so you can tell if an exception is new or recurring.
Parameters not published.
Ask AI
Ask AI about OneUptime
Powered by Claude
Security checks
Static checks derived from the repository 6 hours ago. Heuristics on the source, not an audit.
- ✓README presentThe repository ships documentation.
- ✓License declaredUsable under a stated license.
- ✓Tests presentA test suite exists in the repository.
- ✓Dependencies pinnedA lockfile pins dependency versions.
2 checks have not been run yet.
Releases
README
OneUptime is an open-source observability platform. It checks that your websites, APIs and servers work, collects the logs, metrics and traces your apps send, pages whoever is on call when something breaks, and tells your customers on a status page. It all happens in one product, so the tool that notices a problem is the one that pages your team.
You can use it on OneUptime Cloud or run it on your own servers.
What it does
- Monitoring: Check websites, APIs, ports, DNS and certificates from probes around the world or from inside your own network. OneUptime also watches servers, Kubernetes, Docker, VMware, databases and network devices.
- Incidents and on-call: Work each incident on one timeline. Page whoever is on call by phone, SMS, email, push, Slack or Microsoft Teams, and escalate until someone responds.
- Status pages: Keep customers informed on public or private status pages, and announce scheduled maintenance there.
- Logs, metrics and traces: Send them with OpenTelemetry, search them and build dashboards. Exceptions, profiles, real user monitoring and security events live in the same place.
- Workflows and runbooks: Automate routine work, and connect Jira, ServiceNow and the other tools your team already uses.
- OneUptime AI: Investigates new incidents and alerts and posts a root cause analysis with citations. When you turn fixing on, it can open a pull request with a fix for your team to review.
- API, CLI and Terraform: Automate OneUptime from your scripts and CI, and manage monitors, status pages and on-call as code.
- MCP server (
https://oneuptime.com/mcp): Connect Claude, Cursor, VS Code with GitHub Copilot or any other client of the Model Context Protocol (MCP), and ask it to look into an incident, query logs and traces, acknowledge an alert or set up a monitor.
Everything else is in the docs.
Get started
OneUptime Cloud
Sign up and create a project. The free plan needs no credit card.
Self-host with Docker Compose
You need a Linux server with at least 8 GB of memory, and Docker, Docker Compose, Git and Node.js.
Before step 3, set HOST in config.env to the domain name or IP address people will use to reach OneUptime, without http://. Keep localhost only to try OneUptime on your own computer.
# 1. Clone the release branch
git clone --depth 1 --single-branch --branch release https://github.com/OneUptime/oneuptime.git
cd oneuptime
# 2. Create your config
cp config.example.env config.env
# 3. Start everything
npm start
npm start checks the server, gives every secret in config.env its own random value and starts OneUptime. When it prints OneUptime is up!, open http://localhost, or your server's address, and sign up. The first account becomes the instance's admin.
The Docker Compose guide covers server sizes, HTTPS, backups and updates.
Kubernetes with Helm
For production at scale, or high availability, install the Helm chart. Set host to the domain name or IP address people will use to reach OneUptime:
helm repo add oneuptime https://helm-chart.oneuptime.com/
helm install my-oneuptime oneuptime/oneuptime --set host=oneuptime.example.com
The chart's README covers storage, HTTPS, databases and upgrades.
Community and Enterprise
- Community Edition: Free and open source, with everything above, including SAML and OIDC single sign-on.
- Enterprise Edition: Adds SCIM provisioning, audit logs, team compliance and instance health dashboards, with priority support. Production use needs a license: contact sales.
The docs compare the two editions.
Get help
- Ask questions in the OneUptime community on Slack.
- Report bugs and request features in GitHub issues.
- Report a security vulnerability privately, as SECURITY.md explains.
- Email support@oneuptime.com about anything else.
Contributing
Bug fixes, features, docs and tests are all welcome. The local development guide shows how to run OneUptime from source, and the contributing guidelines explain how pull requests and licensing work. Issues labeled good first issue are a good place to start.
If OneUptime is useful to you, star this repository or sponsor the project.
License
OneUptime is open source under the Apache License 2.0, except for the ee/ directory. That directory holds the Enterprise Edition and is licensed under the OneUptime Enterprise License. The Community Edition image contains no code from ee/. The root LICENSE file sets out this split.
README fetched from the repository 6 hours ago. The original is authoritative.