zixun-github/aisdlc86 installs
spec-test-plan
Use when 需要在 Spec Pack 的 verification 阶段生成或更新 `{FEATURE_DIR}/verification/test-plan.md`(测试计划),并要求严格门禁与可追溯。
How do I install this agent skill?
npx skills add https://github.com/zixun-github/aisdlc --skill spec-test-planIs this agent skill safe to install?
- Gen Agent Trust Hubpass
The skill is a document generation assistant that creates or updates test plans based on markdown templates. It operates entirely on local project documentation without executing code, accessing the network, or handling sensitive credentials.
- Socketpass
No alerts
- Snykpass
Risk: LOW · No issues
- Runlayerpass
1/2 files flagged
What does this agent skill do?
Spec 测试计划(V1:Test Plan)
为当前 Spec Pack 生成/更新测试计划:冻结 范围、策略、环境、准入/准出标准、风险与验证清单,并与 requirements/* 的验收口径保持可追溯一致。
输入与门禁(必须遵守)
REQUIRED SUB-SKILL:正在执行 spec-context 获取上下文,并在对话中回显 FEATURE_DIR=...(允许 (reuse))。
- 必读项目级 memory:
project/memory/product.md、project/memory/tech.md、project/memory/glossary.md;读不到必须写CONTEXT GAP。 - 需求级最小输入:
{FEATURE_DIR}/requirements/solution.md或{FEATURE_DIR}/requirements/prd.md至少其一必须存在;否则停止并写CONTEXT GAP。
输出(落盘)
{FEATURE_DIR}/verification/test-plan.md
强制模板(必须遵守)
test-plan.md必须按模板生成(不得自创结构/字段名):assets/test_plan_template.md
内容要求(最小结构)
test-plan.md 至少包含:
- 执行摘要(待测能力/版本/目标/关键风险)
- 测试范围(In/Out)
- 测试策略(类型:功能/UI/集成/回归/安全…;方法:正向/反向/边界)
- 环境与数据(环境、账号/权限、数据准备方式)
- 准入标准(Entry Criteria)
- 准出标准(Exit Criteria,必须含“阻断交付”的口径)
- 风险与验证清单(Owner/截止/信号/动作)
- 追溯链接(指向
solution/prd的 AC/范围来源;如有#impact-analysis也要链接) - CONTEXT GAP(如有:缺失输入导致的影响)
DoD 自检(V1-DoD)
- 范围 In/Out 明确,且与
requirements/*一致 - 有准入/准出标准,并能明确“什么情况下阻断交付”
- 风险不是清单摆设:每条风险都有最小验证动作(Owner/信号/动作)
- 有追溯入口:至少能回答“本计划依据哪份 AC/范围定义”
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/zixun-github/aisdlc/spec-test-plan">View spec-test-plan on skillZs</a>