AgentSecOps / AgentSecOps/SecOpsAgentKit

[FEATURE] Skills testing framework

未关闭
#2 2 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看
enhancement
主要语言
Python
星标
209
派生
39
PR 合并指标
30 天内没有已合并 PR

描述

## Feature Type
It's important to test skills to ensure that they are validated and triggered properly in different scenarios

## Description

I'm considering an LLM native testing framework like Gherkin or to implement something that is suitable for this use case.

## Use Case

New skill added -> different depth and checklists are validated:
- skill practices (less than 500 lines, refrences, python only scripts etc)
- prompt or skill injection guardrails
- CI triggered
-
Optionally Periodic validations

**Problem it solves:**
ensures new added skills are high quality and consistent throughout the agent kit.

## Willingness to Contribute

- [x] I am willing to submit a PR for this feature
- [x] I can help test this feature
- [x] I can provide domain expertise/guidance

贡献指南

这个仓库没有索引到贡献指南

评估

这个 Issue 还没有评估数据。

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。