K0lmenIA

Design, run and automate tests just by chatting.

K0lmenIA is a team of 28 QA agents for Claude Code. They analyze stories, write test cases, automate with k0lmena and send everything to Xray, QMetry, AIO Tests or Azure DevOps. No coding required.

28specialized agents
4automation types
4test management tools
0tokens to run the suites
how it works

From requirement to result, in three stages

Provide an input (a story, a bug, an API contract, a Jira key or a Figma link), ask for what you need and the right agent produces the result.

  1. Design

    Analyze stories, plan, write manual, BDD and API cases, generate data and write bug reports.

  2. Automation

    Mapper agents write the automation once in k0lmena; then it runs with npm, with zero tokens.

  3. Management

    Cases, cycles and results with evidence in Xray, QMetry or AIO Tests, cases in Azure DevOps, and the closing report.

what you can ask

Ask in plain language. Get ready-to-use artifacts.

You askYou get
"Research the context of PROJ-12"Context sheet and "missing information"
"Build the test plan for HU-001"HTML test plan
"Generate the cases for HU-001"casos-HU-001.xlsx and coverage
"Upload the cases to Xray and create the Sprint 5 cycle"Cases and cycle in Xray
"Automate HU-001's cases in k0lmena".feature, steps and locators
"Build a load test of the login for 20 users"k6, Artillery or JMeter script and HTML report
"Which cases should I run for PR #42?"Risk-based selection and ready command
"Test HU-003 on Chrome, Firefox and Safari"Per-browser matrix
"Fix the tests that failed because of locators"New locators applied and verified
"Write the closing report for HU-001"Go / no-go recommendation

See all 28 agents

agents

A specialist for every QA task

Claude Code picks the agent based on your request. You can also name it: "use the web-mapper to…".

Analysis and design

8 agents

analista-historias, estratega-pruebas, generador-casos-manuales, generador-casos-bdd…

Live execution

10 agents · uses tokens on every run

ejecutor-e2e, ejecutor-api, explorador-web, analista-accesibilidad…

Automation with k0lmena

8 agents · then runs with zero tokens

web-mapper, api-mapper, mobile-mapper, performance-mapper…

Test management

2 agents

gestor-pruebas, publicador-resultados

principles

AI built for serious QA

Don't make things up

If a step, a value or a threshold is missing, the agent flags it and asks. Better an artifact with marked gaps than a complete but invented one.

Traceability

Everything references the story (HU-XXX) and each automated scenario carries the case ID (@CP-001) and the test management key.

Thoughtful coverage

Positive, negative, edge and field validation cases, not just the happy path.

Research before assuming

Before working on a story, it reviews comments, subtasks, epic, linked issues, Confluence and Figma.

Token savings

Anything repetitive is automated once and then runs without agents.

Security

Secrets live in .env, which is never committed; agents never ask for tokens in the chat.

integrations

Connects to your team's tools

Test management, documentation, design, databases and execution. Anything not enabled is marked "Not available", never assumed.

  • Jira
  • Confluence
  • Figma
  • Xray Cloud
  • Xray Server/DC
  • QMetry (QTM4J)
  • AIO Tests
  • Azure DevOps
  • k0lmena TMT
  • PostgreSQL
  • MySQL · MariaDB
  • SQL Server
  • MongoDB
  • Playwright
  • Appium
  • BrowserStack
  • Postman · Newman
  • k6
  • Artillery
  • JMeter
  • OWASP ZAP
  • Lighthouse
  • axe-core

See the architecture

K0lmenIA architecture diagram: Claude Code, agents, skills, MCP connectors, k0lmena and external systems
for companies

K0lmenIA set up for your process

We install it, connect it to your Jira, test management tool and databases, adapt templates to your format and build new agents if your process needs them.

Connectors and credentials

Jira, Confluence, Figma, Azure DevOps, Xray, QMetry or AIO Tests configured securely.

Your own templates and agents

Your case and bug formats, plus custom agents for your team's specific tasks.

Training for manual QA

Your team learns to ask well, review results and automate without coding.

faq

Frequently asked questions

What is K0lmenIA?
K0lmenIA is the AI layer of the k0lmena suite: an open source (MIT) set of 28 QA agents for Claude Code that design, run and automate tests with k0lmena and integrate them with Xray, QMetry, AIO Tests and Azure DevOps.
What do I need to use K0lmenIA?
Claude Code with a Claude account (Pro, Max, Team or Enterprise) or Anthropic API access, Python 3 and Node.js. VS Code is recommended. Everything else (Newman, k6, JMeter, Appium) is only needed if you use it.
What does "zero-token suites" mean?
Mapper agents write the automation once in k0lmena. From then on the suite runs with npm test, without agents and without consuming tokens, as many times as you want.
What if a story is missing information?
Agents look for it in comments, subtasks, epic, linked issues, Confluence and Figma. Anything not found is flagged as missing information (FI-01) with a question for the PO, and marked in cases, .feature files and reports. It is never filled in from what the app happens to do.
Which test management tools does it work with?
Xray Cloud, Xray Server/Data Center, QMetry for Jira (QTM4J) and AIO Tests, with the same commands for all of them. Azure DevOps Test Plans integrates through its official MCP connector. There's also a connector for k0lmena TMT.
Are my credentials safe?
Secrets live in a single .env at the repo root that is never committed. Agents read them from there and never ask for tokens in the chat. Always use test environment credentials.
let's talk

Add AI agents to your QA team.

We help you install K0lmenIA, connect it to your tools and train the team.