dsh-flakefinder
Run the following command in DeepSeek Harness:
dsh plugin install STARDUSTLC666/dsh-flakefinder
Paste the following prompt into your AI chat to install this plugin:
Please install this plugin in DeepSeek Harness using the command dsh plugin install STARDUSTLC666/dsh-flakefinder; the plugin source code is located at https://github.com/STARDUSTLC666/dsh-flakefinder.
About this plugin
In automated testing pipelines, test failures can be particularly frustrating, yet not every failure indicates a genuine code defect. dsh-flakefinder is designed to address this specific challenge, enabling AI Agents to accurately distinguish whether the root cause is "broken code" or simply a "flaky test," thereby preventing wasted debugging efforts.
The plugin boasts robust core capabilities, fully supporting mainstream frameworks such as vitest, jest, pytest, and node:test. By repeatedly executing tests and analyzing results, it can classify them as stable-pass, stable-fail, or flaky. It provides a comprehensive toolchain featuring history queries, quarantine list generation, and write-operation approval gates, ensuring flaky cases are isolated effectively without disrupting the CI pipeline.
This tool is ideal for engineering teams utilizing DeepSeek Harness for intelligent testing, as well as developers looking to enhance CI/CD stability. By decoupling unstable tests from potential code issues, it significantly boosts debugging efficiency and the credibility of test reports, making it an indispensable asset for maintaining high-quality codebases.
Use Cases
- Unit tests failing intermittently and hard to reproduce.
- Using DeepSeek Harness to identify and quarantine flaky tests.
- Generating stability reports to help teams locate code defects.
Best For
- Advanced users of DeepSeek Harness.
- Engineers maintaining test suite quality.
- Teams requiring highly reliable automated testing pipelines.
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.