ReportPortal
ML-powered test reporting: auto-analyze failures across every framework.
About
ReportPortal is an open-source dashboard that collects test results from every framework you run and applies machine learning to the failure pile. Its Auto-Analyzer groups failures by root cause, which the vendor says cuts manual triage effort dramatically, and it adds quality gates so CI/CD pipelines can make automated release decisions on test health. Dashboards show velocity, stability, and coverage trends in real time, and a 2026 release added first-class support for agentic test results, so AI-agent test sessions sit next to classic automation in the same view.
It is Apache-2.0 licensed with 2,038 GitHub stars and commits landing within days of this writing, and it deploys as a Docker stack with agents for all major frameworks. Self-hosting is not trivial (multi-container, Postgres plus OpenSearch), and the ML triage is only as good as your historical data. As a central nervous system for test results at scale, it is one of the more proven open options.
What it can do
- ML-based auto-analysis of test failures
- Real-time test result dashboards
- Quality gates for automated release decisions
- Support for agentic AI test results
- Agents for all major test frameworks
- Unified storage for logs, screenshots, and artifacts
- Trend analysis across launches
Setup
- Deploy ReportPortal with Docker Compose (Postgres + OpenSearch).
- Log in as superadmin and create a project.
- Generate an API key and add the reporter agent for your framework.
- Run your suite, then configure dashboards and quality gates.
Prerequisites
Docker host (approx. 4 vCPU, 8GB RAM); test framework with a ReportPortal reporter agent.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (TIER verified, metrics-v1). Evidence: 2,038 GitHub stars on reportportal/reportportal, Apache-2.0, commits within days of writing (verified via GitHub API); agentic-results support shipped in v26.0.3 per official docs; GitHub Marketplace listing confirms open-source free tier. Strengths: mature open source, real ML triage pipeline, framework-agnostic. Docked for: (1) heavy self-hosted footprint (multi-container Docker stack); (2) ML analysis quality depends on historical run data; vendor triage-reduction figure is provider-claimed. License: Apache-2.0.
★ 2,038 GitHub stars
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 7.0/10
- Maintenance · recency, releases, issue load
- 9.0/10
- Openness · license + docs
- 9.0/10
- Manual review · sanity read + red-flag check
- 8.0/10
- · 2,038 GitHub stars on reportportal/reportportal (GitHub API, 2026-10-16); Apache-2.0
- · Commits within days of writing; agentic-results support shipped in v26.0.3 per official docs
- · GitHub Marketplace listing confirms open-source free tier
- · Heavy self-hosted footprint (multi-container Docker stack: Postgres + OpenSearch)
- · ML analysis quality depends on historical run data; vendor triage-reduction figure is provider-claimed
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.