skills/impact-scan/SKILL.md
# Skill: impact-scan Identify all codebase surfaces affected by an incoming change request. ## Purpose Before writing a spec, we need to know: what already exists that this epic will touch, modify, or depend on? The impact scan produces a blast-radius map that prevents missing cross-cutting concerns. ## Scan Categories ### 1. High Impact Surfaces Files/modules the epic will definitely modify: - Identified by requirement routing - Existing code that will change behavior - Usually tagged `P0`
npx skillsauth add LUAgam/stage-harness skills/impact-scanInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Identify all codebase surfaces affected by an incoming change request.
Before writing a spec, we need to know: what already exists that this epic will touch, modify, or depend on? The impact scan produces a blast-radius map that prevents missing cross-cutting concerns.
Files/modules the epic will definitely modify:
P0Files that depend on directly-impacted modules:
P1Files that may need minor updates:
P2Files/modules that don't exist yet and will be created:
Respect .harness/project-profile.yaml:workspace_mode、risk_level、scan.max_repos_deep_scan、scan.max_files_deep_read_per_scout、scan.max_subagents_wave.
1. Read clarification-notes.md / requirements-draft.md requirements list
2. If workspace_mode is multi-repo:
a. Read .harness/repo-catalog.yaml; build cross-repo-impact-index.json (contracts in interfaces[] first)
b. Deep-scan at most max_repos_deep_scan repos; if more candidates → stop and flag for Lead/user scope convergence
3. For each requirement (and each in-scope repo path):
a. Search codebase for related symbols (Grep) within routed scope only
b. Identify existing implementations
c. Classify impact level
4. Precedent constraint inheritance:
a. Identify existing features/paths that are analogous to the new epic
(same source, same target type, same operation category, or same domain)
b. Search project documentation and code for known limitations,
"not supported" declarations, and prerequisite checks of those
analogous features
c. For each discovered constraint, assess whether it likely applies
to the new epic and record as Risk Flag (category: "precedent-constraint")
5. Build dependency graph (who imports what)
6. Identify test coverage gaps
7. Flag integration points with external systems
8. Score total blast radius: contained / moderate / broad / systemic
Authoritative agent behavior: agents/impact-analyst.md. Templates: templates/cross-repo-impact-index.json, templates/repo-catalog.yaml.
| Score | Description | Risk Modifier | |-------|-------------|---------------| | contained | ≤3 files, isolated module | -1 risk level | | moderate | 4-10 files, single domain | no change | | broad | 11-30 files, multiple domains | +1 risk level | | systemic | >30 files or core infrastructure | force high risk |
This output format is the authoritative contract for impact-scan.md and should stay aligned with commands/harness-clarify.md.
# Impact Scan: <epic-name>
## Blast Radius Summary
- **Blast Radius:** moderate
- **Risk Escalation:** none | +1 medium→high
- <2-3 sentence overview of what will change>
## High Impact Surfaces
| Surface / File | Change Type | Notes | Priority |
|----------------|-------------|-------|----------|
| `src/auth/middleware.ts` | modify | Add new permission check | P0 |
| `src/api/users.ts` | modify | Extend user endpoint | P0 |
## Medium Impact Surfaces
| Surface / File | Reason | Risk | Priority |
|----------------|--------|------|----------|
| `src/api/index.ts` | re-export new handler | low | P1 |
| `tests/auth.test.ts` | existing tests may break | medium | P1 |
## Low / Peripheral Surfaces
| Surface / File | Reason | Priority |
|----------------|--------|----------|
| `docs/auth.md` | docs may need update | P2 |
## New Surfaces (CREATE)
| Path | Purpose | Priority |
|------|---------|----------|
| `src/auth/rbac.ts` | new RBAC module | P0 |
| `tests/rbac.test.ts` | unit tests | P1 |
## Integration Points
| System | Type | Impact |
|--------|------|--------|
| Auth0 | external | token validation contract may change |
| PostgreSQL | database | schema change required |
## Risk Flags
- ⚠️ Auth middleware change affects ALL authenticated routes
- ⚠️ Database migration required — irreversible
## Files NOT Impacted
<list key files explicitly ruled out, to prevent over-scoping>
Invoke skill: impact-scan
Epic: <epic-name>
Requirements: <list from clarification-notes.md>
Output: .harness/features/<epic-id>/impact-scan.md
Optional (multi-repo): .harness/features/<epic-id>/cross-repo-impact-index.json
After scan, impact-analyst reviews and may escalate risk level in project-profile.yaml.
development
在 generate-test-cases 阶段之后执行,逐个验证测试用例并在失败时修复项目代码、重新编译部署、再次验证, 直到通过或达到最大修复次数。覆盖 UI / API / API+UI / 性能测试四个维度,UI 测试通过浏览器真实模拟用户操作并截图, API 测试根据项目代码生成可执行的接口脚本,性能测试调用现有性能/质量技能全量执行。 涉及真实用户登录信息(如手机号+验证码、账号密码、JWT)时必须中断要求用户提供,禁止编造无效凭证。 所有 case 状态变更必须通过 e2e-case-tracker.sh 脚本持久化,确保中途崩溃可恢复、无 case 遗漏。
development
# SKILL: e2e > **核心原则**: > 1. 测试范围跟着本次变动走。后端接口改了,对应的前端流程必须做联调验证;与本次需求无关的功能不测。对于涉及算法、转换准确率等质量敏感型需求,需额外生成专项质量测试。 > 2. **覆盖完整性优先于执行便利性**。不得以"链路复杂"、"需要外部依赖"为由跳过本次变动相关的用例;凡是受变动影响的接口和 UI 流程,都必须生成真实调用/操作用例。 > 3. **UI 测试必须模拟真实用户操作**(定位元素、点击、键入、等待渲染、断言可见文本/状态)。**禁止**将 UI 套件退化为浏览器上下文里的 `page.evaluate(fetch(...))` API 验证——那只是把 API 测试换了执行环境,没有额外价值,不算 UI 测试。 > 4. **通用性**:本 skill 不假设具体业务域,所有规则均以抽象变动面(文件、接口、页面、用户动作)为单位组织,不针对任何特定项目的数据库/领域词汇。 > 5. **E2E 套件必须验证运行时行为**。严禁把"读取源码/配置文件并做字符串/结构匹配"的检查封装成独立 E2E 套件——这类检
tools
# SKILL: deploy ## CLI Bootstrap 在执行任何 `harnessctl` 命令前,先解析本地 CLI 路径: ```bash if [ -z "${HARNESSCTL:-}" ]; then candidates=( "./stage-harness/scripts/harnessctl" "../stage-harness/scripts/harnessctl" "$(git rev-parse --show-toplevel 2>/dev/null)/stage-harness/scripts/harnessctl" ) for candidate in "${candidates[@]}"; do if [ -n "$candidate" ] && [ -x "$candidate" ]; then HARNESSCTL="$candidate" break fi done fi test -n "${HARNESSCTL:-}" && test -x "$H
tools
# SKILL: build ## CLI Bootstrap 在执行任何 `harnessctl` 命令前,先解析本地 CLI 路径: ```bash if [ -z "${HARNESSCTL:-}" ]; then candidates=( "./stage-harness/scripts/harnessctl" "../stage-harness/scripts/harnessctl" "$(git rev-parse --show-toplevel 2>/dev/null)/stage-harness/scripts/harnessctl" ) for candidate in "${candidates[@]}"; do if [ -n "$candidate" ] && [ -x "$candidate" ]; then HARNESSCTL="$candidate" break fi done fi test -n "${HARNESSCTL:-}" && test -x "$HA