dsh-k6
Run the following command in DeepSeek Harness:
dsh plugin install WODE25500/dsh-k6
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install WODE25500/dsh-k6 in DeepSeek Harness to install this plugin; the source is at https://github.com/WODE25500/dsh-k6 .
About this plugin
Writing a k6 load-test script, spawning it, and reading the results used to mean manual CLI calls or hard-coded CI steps. dsh-k6 hands that entire loop to a dsh agent: the agent authors the script, fires a quick 1-VU smoke test to catch syntax errors, then applies real load and pulls a model-readable JSON summary with p95, p99, and error rate—closing the write-run-analyze cycle.
The plugin exposes just four native tools—k6_version, k6_run, k6_scenarios, and k6_smoke—through the lightest possible CLI wrapper with no extra runtime dependencies. The --summary-export JSON is designed for model consumption, so the agent can judge performance thresholds directly without a human parsing terminal logs.
Best for developers and teams already building ops or test workflows in DeepSeek Harness who want k6 load testing folded into an agent automation chain. As an independent community project it builds on grafana/k6 (AGPL-3.0), ships under the MIT license, and drops straight into an existing Harness workflow.
Use Cases
- Agent authors a k6 script, smoke-tests it, then applies real load automatically
- After a run, the agent reads p95, p99, and error rate from JSON to check SLOs
- Fold k6 load testing into an automated regression pipeline inside DeepSeek Harness
Best For
- Developers already building ops or test workflows in DeepSeek Harness
- SREs and backend engineers who want an agent to drive k6 load tests end to end
- Teams that need model-readable performance metrics to automate release decisions
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.