DenisSergeevitch / repo-task-proof-loop
repo-task-proof-loop
Repo-local workflow skill for large coding tasks. Initializes .agent/tasks/TASK_ID artifacts, installs project-scoped Codex and Claude subagents, updates AGENTS.md plus the repo's Claude guide file with the workflow, and runs a spec-freeze → build → evidence → verify → fix loop with fresh-session verification.
Use this skill when the user wants a repeatable, auditable implementation workflow for a non-trivial coding task, especially a feature, refactor, migration, or bug fix that should leave repo-local proof in .agent/tasks/<TASK_ID>/.
All task artifacts created by this workflow must stay inside the repository.
When the examples below mention scripts/task_loop.py, that path is relative to this skill root. Run it while your shell working directory is inside the target repository.
What this skill does
- Initializes a strict repo-local task folder under
.agent/tasks/<TASK_ID>/ - Seeds or updates the required artifact files
- Installs project-scoped Codex and Claude subagent templates into
.codex/agents/and.claude/agents/ - Updates the repo-root
AGENTS.mdCodex baseline plus the repo's Claude guide file (CLAUDE.mdor.claude/CLAUDE.md) with a managed block that explains the workflow - Guides the agent through a strict loop:
- spec freeze - builder implementation - evidence packing - fresh verification - minimal fix - fresh verification again until PASS
SKILL.md