- ROB-1624 Notify Slack when a release workflow fails
- Add allowed_tools to MCP servers to load only some of a server's tools
- feat(alert): add webhook destination
- ROB-616: Migrate AWS MCP to the hosted AWS MCP Server
- Add notify input to eval-benchmarks workflow for quiet manual runs
- ROB-1490 Cleanup repeated Holmes server logs
HolmesGPT by Robusta (CNCF Sandbox)
An open-source agent that investigates production incidents and alerts across Kubernetes, Prometheus, Datadog and other sources to find root causes; a CNCF Sandbox project that runs as a CLI or in-cluster operator.
Use it in: CLI API / SDK Self-hosted Works with: MCP Local models Own API key
On this page
Usage data: latest readings Nov 19 to Nov 21, 2026, last checked Nov 21, 06:26 UTC. Date of each number
What HolmesGPT does
HolmesGPT is an open-source SRE agent for incident root cause analysis from Robusta, first released in 2024. You use it through a command-line tool, an SDK, an API and self-hosted on macOS, Windows and Linux. Its documentation covers MCP servers, your own API key and local models. It can run models from Anthropic, OpenAI, Google, Azure, Amazon Bedrock and OpenRouter and others. It is free and open source (Apache-2.0); you pay only for the models it calls.
Teams use it for: Incident response, DevOps and CI. See examples.
Key facts
| Maker | Robusta (CNCF Sandbox) |
|---|---|
| Type | Open-source SRE agent for incident root cause analysis |
| First released | 2024 |
| License | Apache-2.0 |
| Interfaces | CLI, SDK, API, Self-hosted |
| Platforms | macOS, Windows, Linux |
| Main language | Python |
| Open issues and PRs | 489 |
Pricing
Free and open source (Apache-2.0); bring your own model API key.
Source: official pricing page, checked Oct 6, 2026. Prices change often; confirm before you buy.
Quick answers
- Free to start?Yes
- Open source?Yes (Apache-2.0)
- Runs onmacOS, Windows, Linux
- Where you use itCLI, SDK, API, Self-hosted
- Supports MCP?Yes
- Model providersanthropic, openai, google, azure, amazon, openrouter, any OpenAI-compatible, local
- Feature comparisonAll agents
- Closest alternativeOpenSRE · all alternatives
What's new in HolmesGPT: November 2026 so far
November brought improvements to integrations and logging efficiency, with new webhook capabilities and better tool management for MCP servers.
- Webhook destination added for alert routing.
- MCP servers can now limit which tools are loaded with allowed_tools.
- AWS MCP migrated to hosted server for better reliability.
- Slack notifications now sent when release workflows fail.
1 release this month · summarised from the official release notes · all months
How teams use HolmesGPT
AgentGid articles about HolmesGPT
Get started with HolmesGPT
| PyPI | pip install holmesgpt |
|---|
These commands use the packages tracked on this page. The vendor may recommend a different installer; see the official documentation.
HolmesGPT alternatives
Best HolmesGPT alternatives compared · All AI SRE and DevOps agents
| # | Agent | Gid Score | Downloads 7d | 7d | 30d | Stars | Latest release | Price | |||
|---|---|---|---|---|---|---|---|---|---|---|---|
| 177 |
OpenSRETracer |
28 | 79 | up 61.2% | up 128.6% | 13.7K+127/day | 0.1.2026.11.21today | Free (OSS) | |||
| 236 |
K8sGPTK8sGPT (CNCF Sandbox) |
17 | — | — | — | 8,344+3/day | 0.4.4131d ago | Free (OSS) | |||
| — |
Bits AI SREDatadog |
— | — | — | — | — | — | Usage-based | |||
| — |
ClericCleric |
— | — | — | — | — | — | Free + $700/mo | |||
| — |
incident.io AI SREincident.io |
— | — | — | — | — | — | $25/mo | |||
| — |
Resolve AIResolve AI |
— | — | — | — | — | — | Usage-based | |||
Head to head: HolmesGPT vs OpenSRE · HolmesGPT vs K8sGPT · HolmesGPT vs Bits AI SRE · HolmesGPT vs Cleric · HolmesGPT vs Resolve AI · HolmesGPT vs Traversal
Ratings from people who use HolmesGPT
No ratings yet. Used HolmesGPT? Rate it and say what it is good and bad at.
What changed: recent HolmesGPT releases
7 stable releases in the last 90 days · Monthly summary · Full changelog
- ROB-1291 Multi-tenant Prometheus toolset (Mimir X-Scope-OrgID routed by a Kubernetes label)
- ROB-1525 Reuse the server's cached tool executor in /api/info and the OAuth callback
- Keep an executor rebuilt while a toolset refresh runs
- Check toolsets once in the refresh cycle that builds the executor
- ROB-1365 Stop regenerating the remediation MCP auth token on every render
- Support internal request kinds (oauth_callback, holmes_logs) in tool-call worker
Earlier releases
- ROB-1237 Price Robusta models whose upstream name isn't provider/id
- ROB-1369 Default the auto executor to one conversation at a time
- ROB-1507 - GPT-6.1 Handle multi-choice LLM responses by merging choices
- ROB-801 Refresh litellm cost map when a model lookup misses
- ROB-1291 Per-user Dex auth for the Kubernetes MCP addon, with private CA support
- Weekly Benchmark Results 2026-10-10_21-37
- ROB-357 Remove kube-lineage from Holmes
- ROB-1233 Constrain impact claims in the narrative to observed evidence
- ROB-1107 Bind conversation worker identity to the Conversations row owner
- ROB-1402 Document Baseten Model APIs as a provider
- ROB-1402 Drop the Baseten rate-limits section
- ROB-1104 Reject shell metacharacters in propagated HTTP headers
- Add conversation_link chat field rendered into the system prompt
- Surface MCP structuredContent and drop null optional params
- Add configurable request timeout for New Relic toolset
- Anonymize project slug in MCP toolset test fixture
- Add GPU node diagnostics support to kubernetes-remediation MCP
- Add namespaceScopedRBAC helm value for namespace-scoped Holmes installs
- ROB-1161 Fetch the Supabase api key from relay with caching and local fallback
- Patch all High CVEs (and Mediums >60 days) in the holmes image
- ROB-1258 Live-refresh git-synced skills via first-class skill_repos
- ROB-889 Refresh the realtime JWT before it expires
HolmesGPT prices over time
HolmesGPT popularity and usage data
HolmesGPT was downloaded 4,854 times from PyPI in the week ending Nov 19, 2026, down 0.1% on the week before and up 54.2% over the last 30 days compared with the 30 before. The GitHub repository has 3,949 stars, 304 of them added in the last 30 days. The latest stable release is 0.46.0, published Nov 4, 2026; there was 1 stable release in the past 30 days.
Download the raw daily series: holmesgpt.csv
Gid Score breakdown
Gid Score shows public popularity and activity (usage, GitHub stars, mentions, growth), not the quality of the product. Each component is scored 0 to 100 from public signals; a dash means no data for it. How the Gid Score works.
How fresh these numbers are
| Number | Latest reading | Last checked |
|---|---|---|
| Package downloads | Nov 19, 2026 | Nov 21, 06:24 UTC |
| GitHub stars | Nov 21, 2026 | Nov 21, 06:26 UTC |
We fetch every source once a day. Registries publish downloads and page views a day or two late, so their latest reading trails the check. A reading more than 4 days old is marked out of date; the Gid Score still uses it until a newer one arrives.
HolmesGPT FAQ
How much does HolmesGPT cost?
Free and open source (Apache-2.0); bring your own model API key.
Is HolmesGPT open source?
Yes. HolmesGPT is open source, released under the Apache-2.0 license.
How popular is HolmesGPT?
HolmesGPT was downloaded 4,854 times from PyPI in the week ending Nov 19, 2026, down 0.1% on the week before and up 54.2% over the last 30 days compared with the 30 before. The GitHub repository has 3,949 stars, 304 of them added in the last 30 days. The latest stable release is 0.46.0, published Nov 4, 2026; there was 1 stable release in the past 30 days.
What is the latest version of HolmesGPT?
The latest stable release we track is 0.46.0, published on Nov 4, 2026.
What are the alternatives to HolmesGPT?
The closest alternatives in the same category by Gid Score are OpenSRE, K8sGPT, Bits AI SRE, Cleric.
For the makers of HolmesGPT badges, corrections
Add a live badge to your README or site:
[](https://agentgid.com/agents/holmesgpt/)
[](https://agentgid.com/agents/holmesgpt/)
[](https://agentgid.com/agents/holmesgpt/)
Something wrong or outdated on this page? Send a correction with a link to the official source. More for makers.
Where these numbers come from
PyPI: holmesgpt. GitHub: HolmesGPT/holmesgpt. Site rank: holmesgpt.dev on the Tranco list. See data sources for how each one is collected.
Product facts and pricing were checked against: github.com, pypi.org, cncf.io.