gentle-ai-bench

Agent Skills › Gentleman-Programming/gentle-ai › gentle-ai-bench

gentle-ai-bench

GitHub

用于 gentle-ai 基准测试旅程的编写、验证与调试。涵盖 CI 复现、唯一 ID 管理、Review 模式声明及语义变更检查，确保驱动执行的正确性。

internal/assets/skills/gentle-ai-bench/SKILL.md Gentleman-Programming/gentle-ai

Trigger Scenarios

添加或修改 journey 诊断 bench CI 失败产品语义变更影响旅程

Install

npx skills add Gentleman-Programming/gentle-ai --skill gentle-ai-bench -g -y

More Options

Non-standard path

npx skills add https://github.com/Gentleman-Programming/gentle-ai/tree/main/internal/assets/skills/gentle-ai-bench -g -y

Use without installing

npx skills use Gentleman-Programming/gentle-ai@gentle-ai-bench

指定 Agent (Claude Code)

npx skills add Gentleman-Programming/gentle-ai --skill gentle-ai-bench -a claude-code -g -y

安装 repo 全部 skill

npx skills add Gentleman-Programming/gentle-ai --all -g -y

预览 repo 内 skill

npx skills add Gentleman-Programming/gentle-ai --list

SKILL.md

Frontmatter

{
    "name": "gentle-ai-bench",
    "license": "Apache-2.0",
    "metadata": {
        "author": "Gentleman-Programming",
        "version": "1.0"
    },
    "description": "Trigger: bench, journey, journeys, driven mode, gentle-ai-bench, journey corpus, j-numbers, bench axis. Author and verify gentle-ai bench journeys; go test .\/bench never proves driven execution."
}

Activation Contract

Load when touching bench/ in gentle-ai, adding or changing a journey, changing a product semantic a journey might pin, or diagnosing a bench failure in CI's Unit Tests job.

Hard Rules

go test ./bench validates corpus declarations only. It does NOT execute journeys. The only driven proof is building the harness and the product binary and running the harness against it; a green go test ./bench claims nothing about execution.
Reproduce CI, do not guess invocations: read the Unit Tests step in .github/workflows/ci.yml and copy its exact build and gentle-ai-bench run --binary ... commands. Use --only <journey-id> to drive one journey.
Journey IDs are unique across every journeys_*.go file. The collision guard fails loudly naming both files; pick an unused ID by reading the corpus, never reuse a retired one.
Every journey declares Review: — reviewOptedIn (the runner enables receipt-driven development globally before the first step, uncounted, and fails the journey if the switch does not come on) or reviewUntouched (its subject IS the switch, or it has nothing to do with reviews). The declaration is mandatory; validateCorpus fails the run without it. Never let a journey inherit the product's default: reviews are opt-in, and a journey that assumed otherwise measures a review-refused flow while still reporting completed.
Every execute transition must carry a runnable command; the dead-execute guard fails the run otherwise.
When a ratified product semantic changes, grep the corpus for journeys pinning the OLD behavior before shipping. The corpus is a second test surface beyond unit tests; a journey asserting the defect keeps the defect green.
dead_end prints n/a unless the run actually measured one. Never fabricate a value to move the column.
A by_design exemption costs a shape from the closed vocabulary plus a verified quote of the product's own next-action text. If the quote no longer tells the operator what to do, it is a defect wearing an exemption.
Prefer a NEW journeys_*.go file when the shared ones are owned by open PRs; bump the core journey-count pin in the same change.

Execution Steps

Read the corpus area you touch and the CI invocation before writing.
Author or adapt the journey; update its title, step names, and comment to say WHY the expectation holds (cite the issue or ratified decision).
Run go test ./... in bench/ for declarations, THEN the driven harness for execution; both results go in the PR body.
On semantic changes, list the journeys you checked for stale pins.