harness-benchmark

Propose project-specific benchmark scripts and pass/fail thresholds for critical paths.

Updated Jul 29, 2026
One-click install
npx skills add https://github.com/MaiconGambini/opencode-harness-guide --skill harness-benchmark-maicongambini
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: harness-benchmark
Source: https://github.com/MaiconGambini/opencode-harness-guide/tree/main/skills/harness-benchmark
Command: npx skills add https://github.com/MaiconGambini/opencode-harness-guide --skill harness-benchmark-maicongambini

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams often lack performance regression checks tied to their actual critical user flows, making slowdowns invisible until users complain. This Skill guides the discovery of core flows and produces a concrete benchmark proposal with explicit thresholds before any script is written. ## Core Features & Use Cases - Critical Path Discovery: Identifies core user or system flows from docs, routes, commands, or tests, plus any existing benchmark, load test, Lighthouse, or timing scripts. - Threshold Definition: Requires explicit pass/fail thresholds and records environment assumptions for repeatable local checks. - Approval-Gated Output: Produces a structured Benchmark Proposal (critical path, metric, threshold, command, data setup) and never writes benchmark scripts without approval. - Use Case: Before optimizing an API endpoint, use this Skill to propose a benchmark script targeting the checkout flow with a 500ms p95 threshold, then get approval before generating scripts/benchmark.ps1 or scripts/benchmark.sh. ## Quick Start Ask the agent to propose a benchmark for the most critical user flow in this project with explicit pass/fail thresholds.

Frequently Asked Questions about harness-benchmark

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a benchmark for my project's critical path?

Start by identifying core user or system flows from docs, routes, commands, or tests, then pick one critical path. Define an explicit pass/fail threshold and metric, and get approval before writing any benchmark script such as scripts/benchmark.ps1 or scripts/benchmark.sh.

What should a performance benchmark proposal include?

A benchmark proposal should state the critical path, the metric being measured, the pass/fail threshold, the command to run it, and the data setup required. It should also record environment assumptions so results stay repeatable.

Does this Skill write benchmark scripts automatically?

No. The Skill only produces a benchmark proposal and explicitly requires approval before creating any scripts. This gate ensures thresholds and critical paths are reviewed before automation is added.

Should I use local benchmarks or external load testing services?

Prefer repeatable local checks first, since they are easier to run in CI and debug. External services can come later once local thresholds and environment assumptions are established.

When is a benchmark proposal not the right approach?

If the project has no identifiable critical flow or no measurable metric, a benchmark adds little value. It also does not replace full load testing for high-traffic capacity planning scenarios.