@sofagent/ab-test
Advanced tools
sofagent A/B 测试框架——对比实验、指标显著性分析、自动 promote 决策。
# 运行 A/B 对比测试（从 golden-set YAML 加载测试用例）
sofagent-ab-test run --current <skill-path> --candidate <skill-path> [options]
# 选项
sofagent-ab-test run --current <path> # 当前版本 Agent Skill 路径
sofagent-ab-test run --candidate <path> # 候选版本 Agent Skill 路径
sofagent-ab-test run --eval-set <path> # golden-set YAML 路径（默认从 @sofagent/eval 的 golden-set.yaml 加载）
sofagent-ab-test run --threshold <n> # 晋升阈值（默认 2）
# 测试用例不足时报错退出（不再使用硬编码默认用例）
运行结果持久化到 data/ab-test/latest.json（覆盖写）。
v1.2.4 起，ab-test CLI 的 run 子命令从 @sofagent/eval 的 golden-set YAML 加载测试用例，
与 eval 共享同一套 input + expected。ab-test 的 runner 是 Agent Skill 对比（current vs candidate），
不是 audit runRules——ab-test 只复用 golden set 作为 TestCase 来源。
decidePromotion() — 对比 A/B 方案指标，连续 2 轮胜出 → auto promotepersistABTestResult() — 持久化 A/B 测试结果到 latest.jsonDEFAULT_SCORE_WEIGHTS — 默认评分权重ABConfig / ABTestResult / PromotionDecision / ScoreWeights@sofagent/eval / @sofagent/core / @sofagent/inject / @sofagent/orchestratorFAQs
sofagent A/B 测试框架 — 对比实验 / 指标显著性 / 实验报告
The npm package @sofagent/ab-test receives a total of 63 weekly downloads. As such, @sofagent/ab-test popularity was classified as not popular.
We found that @sofagent/ab-test demonstrated a healthy version release cadence and project activity because the last version was released less than a year ago. It has 1 open source maintainer collaborating on the project.
Security News
The PHP Foundation says Chainguard broke a security embargo ahead of critical RCE patches, as leaks disrupted the coordinated release.
Security News
PHP releases emergency patches for two critical RCE vulnerabilities as agencies urge organizations to take unpatched sites offline.
Threat Research
/Security News
A supply chain attack compromised LiteLLM 1.107.0 on PyPI and Docker, spreading malware that steals cloud credentials, developer secrets, and AI provider keys.