Agentic QE Fleet plugin
QA agents, skills and an MCP server: test generation, coverage gaps, flaky tests, security scans and chaos testing
- 519 ★ on the repository
- Last commit
- MIT
Broad access
Why this level
- Agents run tests and arbitrary commands in the project
- Chaos and pentest agents can inject faults and probe test environments
- Creates a local database and agent configuration in the project
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/agentic-qe-fleetDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Agents run tests and arbitrary commands in the project
- Chaos and pentest agents can inject faults and probe test environments
- Creates a local database and agent configuration in the project
This is third-party code. Review the repository files before installing.
What it does
Agentic QE installs into a project as an npm package and connects an MCP server, dozens of specialized QA agents and a testing skill library to your coding agent. Agents generate unit, integration, property-based and BDD tests for your framework, find untested code with risk scoring, analyze root causes of flaky tests and run security and resilience checks. A coordinator distributes work across agents and turns results into a release recommendation. For Claude Code there is a slimmer plugin with a subset of agents, slash commands and skills.
Who it is for. For QA engineers and development teams who want to hand systematic test work to an agent.
Good fit when
- You need to quickly add tests to an existing service
- You want to find the riskiest uncovered code
- CI has many flaky tests
- You want a TDD loop with code and security review
Not a fit when
- A small project where a plain agent is enough
- You cannot add heavy npm dependencies and a local memory database to the project
Example request
Find coverage gaps in src/ and prioritize them by riskLimitations
The full aqe init setup is heavy: native modules, a local pattern database and many agents. On Windows without build tools, vector search falls back to brute force and is unsuitable for large codebases. The Claude Code plugin works only in Claude Code and includes a subset of features. Your agent still needs access to its model provider.
How to disable. Uninstall the agentic-qe-fleet plugin via /plugin; for the full setup run npm uninstall -g agentic-qe and remove files created by aqe init, including the .agentic-qe folder.
FAQ
Plugin or aqe init?
The plugin is for a quick start in a single Claude Code project. aqe init installs the full agent and skill set with a persistent learning database and supports other agents like Cursor, Copilot and Cline.
Which agents are supported?
Claude Code, GitHub Copilot, Cursor, Cline, OpenCode, Kiro, Kilo Code, Roo Code, Codex CLI, Windsurf and Continue.
Discussion
No questions or reviews yet.
Related
PluginsA self-improving QA agent: natural-language tests for web and mobile that self-heal and remember runs
MCP server, agent and skills for end-to-end UI tests where the agent drives the computer through vision on cloud machines
A skill that tests your agent's code in a real browser: reads your changes, builds a test plan and runs it through Playwright with your own cookies
CLI, MCP server and typed Node.js API so an agent can verify its changes in a live mobile app on iOS, Android and more
An MCP server that spins up several browser agents to test a site for broken links, UI bugs and accessibility issues
An MCP server that checks whether the agent is really done: drives the real app and returns pass, fail or couldn't tell with the file and line
README in short
The README covers two setups: a global npm package with aqe init that detects the stack and configures MCP, and a Claude Code plugin with a reduced set. Agents are grouped by domain: test generation and execution, coverage, quality assessment, defects, requirements, security, contracts, visual and accessibility checks, chaos and performance, and enterprise integrations. Skills are labeled with trust tiers by verification level. It includes example prompts and Windows-specific notes. MIT licensed.