القائمة
Find and remove tech debt (redundant/duplicated code), run linters, and ensure code quality in recent changes
يُحدَّد Prompt الذي يراجع المصدر أولًا بشكل افتراضي. يمكنك التبديل إلى أمر مباشر أو تنزيل نسخة محلية.
اقرأ SKILL.md وأي ملفات مرافقة يعرضها SkillsMP قبل أن تقرر التثبيت.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
يتجاوز الأمر المباشر Prompt المخصّص للمراجعة. افحص المصدر قبل تشغيله.
npx skills add https://github.com/meta-pytorch/torchrec --skill techdebtيبقى الأمر في سطر واحد. مرّر أفقيًا لمراجعته كاملًا قبل النسخ.
تفضّل نسخة محلية؟ نزّل الملفات المتاحة حاليًا لدى SkillsMP.
Generate tests for TorchRec source files with correct patterns (unit, distributed, hypothesis), proper BUCK targets, and test utilities. Use when asked to generate tests, add test coverage, or write tests for a module.
Review TorchRec pull requests and diffs for distributed correctness, sharding safety, backward compatibility, and test coverage. Use when reviewing PRs, diffs, or when asked to review code changes.
Investigate and explain TorchRec planner sharding statistics output, especially how HBM storage is computed per table and per rank. Use when the user asks about sharding stats, storage breakdown, or memory estimation.
عرض SKILL.md
استنادا إلى تصنيف SOC المهني
| name | techdebt |
| argument-hint | ["path or instructions"] |
| description | Find and remove tech debt (redundant/duplicated code), run linters, and ensure code quality in recent changes |
| allowed-tools | AskUserQuestion, Write, Read, Edit, Bash, Grep, Glob, Task |
You are a tech debt cleanup specialist. Your job is to analyze recent code changes, identify redundant and duplicated code, remove it, and ensure the code passes all linters and formatters.
Find changed files by running:
sl status
sl diff --stat
If no uncommitted changes exist, check recent TorchRec commits (marked with [torchrec] tag):
sl log -l 50 -T "{node|short} {desc|firstline}\n" | grep "\[torchrec\]" | head -5
List the files that have been modified and will be analyzed.
For each changed file, look for:
Redundant Code:
Duplicated Code:
Code Quality Issues:
# pyre-strict)Before making changes, present a summary to the user:
## Tech Debt Analysis Summary
### Files Analyzed: [N]
### Issues Found:
**High Priority (in recently changed code):**
1. [file:line] - [description of issue]
2. ...
**Medium Priority (adjacent to changed code):**
1. [file:line] - [description of issue]
2. ...
**Low Priority (elsewhere in file):**
1. [file:line] - [description of issue]
2. ...
### Proposed Changes:
- Remove N unused imports
- Delete N lines of dead code
- Consolidate N duplicated patterns
- Fix N code quality issues
Ask the user: "Would you like me to proceed with these changes? (You can also specify which categories to address)"
After user approval:
Important Guidelines:
After making changes, run the appropriate linters based on file types:
For Python files:
arc lint -a <changed_files>
For all files (general):
arc lint
If lint errors are found:
arc lint -aFor Python files, run Pyre type checking:
arc pyre check-changed-targets
If type errors are found, fix them and re-run until clean.
## Tech Debt Cleanup Complete
### Changes Made:
- [file]: Removed N unused imports, deleted M lines of dead code
- [file]: Extracted duplicated pattern into helper function `foo()`
- ...
### Linting Status: ✓ All checks passed
### Files Modified: [list]
If the user provides $ARGUMENTS: