Show HN: Sentinel – open-source QA agent that reads your code before it clicks

4 hours ago

Sentinel is an open-source QA agent that reads the codebase to understand business flows and performs end-to-end testing across frontend and backend.
It autonomously derives critical business flows from the code, such as reservation lifecycle, group bookings, and payment processes, similar to a human QA engineer.
During testing, Sentinel interacts with both UI and API, catching backend bugs that UI-only tests miss, like state-machine inconsistencies and API/UI disagreements.
The agent uses a pipeline: recon pass extracts code structure, model derives flows, Playwright drives browser, and API tools check server state, with multiple attempts per flow.
It includes a vision pass for design review, scoring visual hierarchy, spacing, typography, and WCAG contrast issues.
Sentinel handles apps without login forms by injecting a wallet interface for web3 apps, using unfunded burners to prevent real transactions.
The stack is minimal (~2,500 lines), runs on a schedule, and includes sibling agents for code review, docs sync, and brain sync.
Current limitations include variability in exploration, config complexity for booting stacks, and depth/time/cost trade-offs.
Sentinel is MIT-licensed, available on GitHub, with configurable model providers and ready-to-use examples for real-world testing.

Hasty Briefsbeta