home / skills / levnikolaevich / claude-code-skills / ln-403-task-rework

ln-403-task-rework skill

safe

/ln-403-task-rework

This skill reworks a single To Rework task by applying reviewer feedback and restoring it to To Review.

npx playbooks add skill levnikolaevich/claude-code-skills --skill ln-403-task-rework

Review the files below or copy the command above to add this skill to your agents.

Files (2)

SKILL.md

3.0 KB

---
name: ln-403-task-rework
description: Fixes tasks in To Rework and returns them to To Review. Applies reviewer feedback only for the selected task.
---

> **Paths:** File paths (`shared/`, `references/`, `../ln-*`) are relative to skills repo root. If not found at CWD, locate this SKILL.md directory and go up one level for repo root.

# Task Rework Executor

Executes rework for a single task marked To Rework and hands it back for review.

## Purpose & Scope
- Load full task, reviewer comments, and parent Story; understand requested changes.
- Apply fixes per feedback, keep KISS/YAGNI, and align with guides/Technical Approach.
- Update only this task: To Rework -> In Progress -> To Review; no other tasks touched.

## Task Storage Mode

| Aspect | Linear Mode | File Mode |
|--------|-------------|-----------|
| **Load task** | `get_issue(task_id)` | `Read("docs/tasks/epics/.../tasks/T{NNN}-*.md")` |
| **Load review notes** | Linear comments | Review section in task file or kanban |
| **Update status** | `update_issue(id, state)` | `Edit` the `**Status:**` line in file |

**File Mode transitions:** To Rework → In Progress → To Review

## Workflow (concise)
1) **Receive task:** Get task ID from orchestrator (ln-400); read task (Linear: get_issue; File: Read task file), review notes, parent Story.
2) **Plan fixes:** Map each comment to an action; confirm no new scope added.
3) **Implement:** Follow task plan/checkboxes; address config/hardcoded issues; update docs/tests noted in Affected Components and Existing Code Impact.
4) **Quality:** Run typecheck/lint (or project equivalents); ensure fixes reflect guides/manuals/ADRs/research.
5) **Root Cause Analysis:** Ask "Why did the agent produce incorrect code?" Classify: missing context | wrong pattern | unclear AC | gap in docs/templates. If doc/template gap found → update the relevant file (guide, template, CLAUDE.md) to prevent recurrence.
6) **Handoff:** Set task to To Review (Linear: update_issue; File: Edit status line); move it in kanban; add summary comment referencing resolved feedback + root cause classification.

## Critical Rules
- Single-task only; never bulk update.
- Do not mark Done; only To Review (ln-402 decides Done).
- Keep language (EN/RU) consistent with task.
- No new tests/tasks created here; only update existing tests if impacted.
- **Do NOT commit.** Leave all changes uncommitted — ln-402 reviews and commits with task ID reference.

## Definition of Done
- Task and review feedback fully read; actions mapped.
- Fixes applied; docs/tests updated as required.
- Quality checks passed (typecheck/lint or project standards).
- Root cause classified (missing context | wrong pattern | unclear AC | doc gap); doc/template updated if gap found.
- Status set to To Review; kanban updated; summary comment with fixed items + root cause.

## Reference Files
- **[MANDATORY] Problem-solving approach:** `shared/references/problem_solving.md`
- Kanban format: `docs/tasks/kanban_board.md`

---
**Version:** 3.0.0
**Last Updated:** 2025-12-23

Overview

This skill executes rework for a single task that is currently marked To Rework and returns it to To Review after applying reviewer feedback. It focuses only on the selected task, follows project guides and ADRs, and enforces quality checks without committing changes. The skill preserves scope discipline and records a clear root-cause classification for the failure and remedial action.

How this skill works

The skill loads the full task, its review comments, and the parent Story from the project storage (Linear API or file mode), maps each reviewer comment to a concrete action, and implements fixes within the task boundaries. It runs lint/type checks or project-equivalent validations, updates affected docs or tests if directly impacted, classifies the root cause, and updates the task status To Rework → In Progress → To Review while adding a summary comment. Nothing is committed; changes remain uncommitted for the next orchestrator to review and commit.

When to use it

A single task is marked To Rework and requires changes requested by reviewers.
You need to apply only the reviewer feedback for one task without changing scope.
Rework involves code/config/docs in the task’s Affected Components or Existing Code Impact.
A root-cause classification and brief template or guide update may prevent recurrence.
You must leave changes uncommitted for a subsequent review/commit step.

Best practices

Map each review comment to a minimal actionable change; avoid adding new features.
Run project lint/typecheck and relevant unit checks locally before status update.
Update only the task and directly impacted docs/tests; do not create new tasks.
Classify the root cause (missing context | wrong pattern | unclear AC | doc gap) and update templates/guides when appropriate.
Keep task language consistent with the original task (EN/RU) and document the resolved items clearly in the summary comment.

Example use cases

Fix a failing code example and update a linked README snippet noted in review comments.
Adjust configuration or hardcoded values referenced by reviewers and re-run linters.
Address a UX copy change requested in the task’s review notes and update the task file.
Apply a small test fix flagged in Existing Code Impact and re-run the test suite.
Classify a pattern mismatch as wrong pattern and add a short note to the coding guide to prevent repeats.

FAQ

Will the skill batch-update multiple tasks?

No. It strictly operates on a single selected task and will never update other tasks.

Does it commit changes after rework?

No. All edits are left uncommitted so the orchestrator (ln-402) can review and commit with the task ID.