원클릭으로
clean-tree-precondition
Every editing agent should start on a clean working tree so parallel work never collides.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Every editing agent should start on a clean working tree so parallel work never collides.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | clean-tree-precondition |
| description | Every editing agent should start on a clean working tree so parallel work never collides. |
Before an agent creates or edits files, the working tree should be clean — no uncommitted changes from a previous task. Otherwise two pieces of work edit over each other and produce a tangled, unreviewable diff.
git status --porcelain # empty output = clean
git status --short lists staged/untracked).commit-message skill). If it is unrelated, stash it deliberately — never
edit on top of it.git status --porcelain is empty.The post-commit hook refreshes the generated context/index (CURRENT_STATE, RECENT_CHANGES, FILE_REGISTRY, CODE_MAP) after each commit, so they may show as dirty. That churn is expected — fold it into the next commit; it is not a reason to stop work.
Do not begin a task on a dirty tree of unrelated work. A clean start is what makes sequential delegation (and later, worktrees) safe.
Keep commits atomic — split a mixed change set into scoped, reviewable commits.
The one-time procedure to take an exported project to live — understand it, confirm the name/goal/stack with the user, run the deterministic `aspis bootstrap`, enrich the judgment files (AGENTS.md, ARCHITECTURE, context), verify, and let the package self-clean. Followed by the bootstrap agent.
Keep per-subsystem architectural intent current through the planning loop — read before designing, record an impact report on change, confirm with the user, apply a dated update, and verify the build against approved intent.
Audit a plan's task dependency graph for structural integrity — circular dependencies, missing prerequisites, orphan tasks, and dependency classification (hard/soft/warning). Produces a pass/warn/fail audit report per dependency so planners catch graph errors before build starts.
Verify hooks ran, no secrets, protected paths untouched, and commit message valid before committing. Owned by the reviewer.
Prove the catalog regenerates the live runtime byte-for-byte; refuse export on any mismatch.