code-reviewer

النجوم٦

التفرعات٠

آخر تحديث١ يونيو ٢٠٢٦ في ٠٧:٠٠

Review repository code changes with a findings-first focus on correctness, regressions, compatibility, and missing validation. Use for code review, risk review, or model adaptation review requests.

التثبيت

التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.

المصدر

houmo-ai

houmo-ai/houmo-xh2modelzoo

فتح مستودع GitHub عرض مستودعات المنشئ

تنزيل

المهن ذات الصلةSOC

استنادا إلى تصنيف SOC المهني

محللو ضمان جودة البرمجيات والمختبرونمهن الحاسوب والرياضيات·SOC 15-1253

SKILL.md

readonly

name	code-reviewer
description	Review repository code changes with a findings-first focus on correctness, regressions, compatibility, and missing validation. Use for code review, risk review, or model adaptation review requests.
license	Apache-2.0
metadata	{"author":"Lihui","version":"3.0"}

Code Reviewer

Use this skill when the user asks for code review, risk review, regression review, or quality review.

Scope

Review only. Do not edit code unless the user explicitly switches from review to implementation.
Prioritize real defects and behavior risks over style-only comments.
Keep the review grounded in the requested scope. If the user names files, models, or directories, do not expand the scope without saying so.

Review Priorities

Review in this order:

Correctness and behavioral regressions
Compatibility with existing APIs, configs, and model registration
Cache, shape, dtype, and sequence-length contracts
Reliability, error handling, and edge cases
Missing or weak validation
Maintainability and readability

Repository-Specific Checks

For code under xhmodel_merak/xh_llm/models, check registration, wrapper replacement, cache flow, and output semantics.
For CLI changes, check command routing, argument compatibility, and help behavior.
For config or example changes, check that names, paths, and parameters still match the code they exercise.
If the review is about model adaptation compatibility, also read .github/skills/xhquant_llm_reviewer/SKILL.md.

Model Review Routing

When the user asks to review a specific adapted model, default review targets are:

xhmodel_merak/xh_llm/models/<model_name>/
examples_merak/llm/<model_name>/ if it exists

When the user asks for a broad LLM review, enumerate model directories under xhmodel_merak/xh_llm/models/ and review them one by one instead of giving a vague aggregate answer.

Output

Return findings first, ordered by severity.

For each finding include:

file or symbol
concrete issue
impact
minimal fix direction

Then include:

open questions or assumptions
residual risk or validation gaps
a short overall conclusion

If there are no findings, say that explicitly and still mention residual risks or unverified areas.

المزيد من هذا المستودع

نفس المستودع

migrate-xhquantllm-to-xh2modelzoo

houmo-ai/houmo-xh2modelzoo

将 xhquant_llm 模型家族迁移到 xh2modelzoo 的 LLMConverter 架构，并完成 converter/convert_config、architecture 注册、examples 导出脚本与 hmonnx 验证闭环。当用户要求把已有 xhquant_llm 模型同步到 xh2modelzoo、补齐 xh2modelzoo 缺失模型支持、排查迁移后导出或推理失败时使用。

2026-06-016

xhquant-llm-example-scripts

houmo-ai/houmo-xh2modelzoo

Build or update example scripts for adapted LLM and VLM models, covering export, generate, and optional quant flows in the style used by this repository.

2026-06-016

xhquant-llm-reviewer

houmo-ai/houmo-xh2modelzoo

Review whether an adapted transformers LLM or VLM model is compatible with the xhmodel_merak architecture. Focus on wrapper registration, graph_forward legality, cache flow, and output contracts.

2026-06-016

xhquant-transformers-to-graph

houmo-ai/houmo-xh2modelzoo

Convert a Hugging Face transformers text model or multimodal model into an xhquant-callable and executable graph by following the wrapper, graph_forward, frontend, HMONNX, and runtime-contract patterns used by qwen3_legacy, qwen3_vl, and qwen3_5.

2026-06-016

name	code-reviewer
description	Review repository code changes with a findings-first focus on correctness, regressions, compatibility, and missing validation. Use for code review, risk review, or model adaptation review requests.
license	Apache-2.0
metadata	{"author":"Lihui","version":"3.0"}

Code Reviewer

Use this skill when the user asks for code review, risk review, regression review, or quality review.

Scope

Review only. Do not edit code unless the user explicitly switches from review to implementation.
Prioritize real defects and behavior risks over style-only comments.
Keep the review grounded in the requested scope. If the user names files, models, or directories, do not expand the scope without saying so.

Review Priorities

Review in this order:

Correctness and behavioral regressions
Compatibility with existing APIs, configs, and model registration
Cache, shape, dtype, and sequence-length contracts
Reliability, error handling, and edge cases
Missing or weak validation
Maintainability and readability

Repository-Specific Checks

For code under xhmodel_merak/xh_llm/models, check registration, wrapper replacement, cache flow, and output semantics.
For CLI changes, check command routing, argument compatibility, and help behavior.
For config or example changes, check that names, paths, and parameters still match the code they exercise.
If the review is about model adaptation compatibility, also read .github/skills/xhquant_llm_reviewer/SKILL.md.

Model Review Routing

When the user asks to review a specific adapted model, default review targets are:

xhmodel_merak/xh_llm/models/<model_name>/
examples_merak/llm/<model_name>/ if it exists

When the user asks for a broad LLM review, enumerate model directories under xhmodel_merak/xh_llm/models/ and review them one by one instead of giving a vague aggregate answer.

Output

Return findings first, ordered by severity.

For each finding include:

file or symbol
concrete issue
impact
minimal fix direction

Then include:

open questions or assumptions
residual risk or validation gaps
a short overall conclusion

If there are no findings, say that explicitly and still mention residual risks or unverified areas.