Requesting Code Review
HezaoHezao/poirot
Pre-commit review: security scan, quality gates, auto-fix. An agent skill from HezaoHezao/poirot.
Runs a light convention check on one finished spec-driven task, choosing checks by task type and ending in pass, pass-with-notes or needs-fix.
SKILL.md written in Chinese; this summary is our English description.
Install Task-Level Code Review
$ npx skills add dataelement/bisheng --skill task-review -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install dataelement/bisheng task-review --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/dataelement/bisheng.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/task-review .claude/skills/task-review && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "task-review" agent skill from https://github.com/dataelement/bisheng/tree/main/.claude/skills/task-review into .claude/skills/task-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "task-review", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/dataelement/bisheng/tree/main/.claude/skills/task-reviewType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add dataelement/bisheng --skill task-review -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install dataelement/bisheng task-review --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/dataelement/bisheng.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/task-review .agents/skills/task-review && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "task-review" agent skill from https://github.com/dataelement/bisheng/tree/main/.claude/skills/task-review into .agents/skills/task-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "task-review", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add dataelement/bisheng --skill task-review -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install dataelement/bisheng task-review --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/dataelement/bisheng.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/task-review .cursor/skills/task-review && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "task-review" agent skill from https://github.com/dataelement/bisheng/tree/main/.claude/skills/task-review into .cursor/skills/task-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "task-review", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/dataelement/bisheng.git --path .claude/skills/task-review--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add dataelement/bisheng --skill task-review -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install dataelement/bisheng task-review --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/dataelement/bisheng.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/task-review .gemini/skills/task-review && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "task-review" agent skill from https://github.com/dataelement/bisheng/tree/main/.claude/skills/task-review into .gemini/skills/task-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "task-review", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install dataelement/bisheng task-reviewInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add dataelement/bisheng --skill task-review -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/dataelement/bisheng.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/task-review .github/skills/task-review && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "task-review" agent skill from https://github.com/dataelement/bisheng/tree/main/.claude/skills/task-review into .github/skills/task-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "task-review", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add dataelement/bisheng --skill task-review -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install dataelement/bisheng task-review --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/dataelement/bisheng.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/task-review .opencode/skills/task-review && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "task-review" agent skill from https://github.com/dataelement/bisheng/tree/main/.claude/skills/task-review into .opencode/skills/task-review/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "task-review", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
task-review- GitHub stars
- 12k
- Token cost
- ~652 tokens
- SKILL.md length
- 166 words
- Files
- 2 (incl. references)
- Skills in repo
- 10
- Repo updated
- First seen
- Licence
- Apache-2.0
At a glance
Runs a light convention check on one finished spec-driven task, choosing checks by task type and ending in pass, pass-with-notes or needs-fix.
- Works in 6 steps: 解析参数 + 收集变更范围 → 判断任务类型，选择检查子集 → 按检查清单执行检查 → …
- Checking a just-finished spec-driven task before ticking it off
- SKILL.md covers 调用方式, 审查流程 and 错误处理
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
Written in Chinese, this is the first level of a two-level review. After a task from a feature's `tasks.md` is done, you run `/task-review` with the feature directory and task ID. The agent reads the task's metadata, such as its type, target files, prerequisites, paired test or implementation task and covered acceptance criteria, then reads the target files directly instead of relying on git diff, so architecture and coding rules are enforced before violations pile up for the feature-level review.
Each task is classed as test, implementation, infrastructure or worker from its file paths and description, and each class gets its own subset of the checklist in `references/task-checklist.md`. Cross-checks cover whether declared files exist, whether the task strayed beyond them, and whether the paired test task and prerequisites are ticked. The report ends in PASS, PASS_WITH_NOTES or NEEDS_FIX, with one re-review allowed after fixes, and a missing directory, tasks file or task ID stops the run with an error.
When your agent uses it
- Checking a just-finished spec-driven task before ticking it off
- Catching architecture-layer or naming violations early, at task level
- Re-reviewing a task after fixing high-severity findings
Example prompts
- “/task-review features/v2.5.0/004-rebac-core T003”
- “Review task T007 in features/v2.5.0/007-resource-permission-ui against the task checklist.”
- “Run the task review for T003 again now that I have fixed the HIGH issues.”
Requirements
- A feature directory that contains a `tasks.md` file
Workflow steps
6 steps, taken from the step headings in SKILL.md.
What it can do on your machine
Read from SKILL.md and the folder at commit ed3b461. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Task-Level Code Review loads about 652 tokens when it runs, and up to ~1.8k if it reads all its reference files. Until then it costs about 46 tokens; SKILL.md has 166 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from dataelement/bisheng at commit ed3b461, republished under its Apache-2.0 licence (© dataelement). 166 words, ~652 tokens.
Download SKILL.mdSave it as
.claude/skills/task-review/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.- name
- task-review
- description
- L1 任务级代码审查。在每个任务完成后执行轻量级约定合规检查， 确保架构红线和编码约定在任务级别被守住，不让违规累积到特性级审查（L2）才发现。 用法：/task-review <feature_dir> <task_id> TRIGGER when: 用户完成了一个 SDD 任务（实现或测试），或者用户使用 /task-review 命令。
Task Review Skill（L1 任务级审查）
调用方式
/task-review <feature_dir> <task_id>例：
/task-review features/v2.5.0/004-rebac-core T003
/task-review features/v2.5.0/007-resource-permission-ui T007审查流程
Step 1: 解析参数 + 收集变更范围
验证参数：
feature_dir必须存在且包含tasks.mdtask_id必须匹配 tasks.md 中的某个任务（格式：T001、T003等）- 若参数缺失或无效，报告错误后停止
从
<feature_dir>/tasks.md中读取指定任务的元数据：- 任务类型（测试 / 实现 / 基础设施 / Worker）
- 目标文件列表
- 前置依赖
- 配对任务（测试↔实现）
- 覆盖 AC 标注（测试任务）
读取任务声明的所有目标文件内容（直接读取文件，不依赖 git diff）
Step 2: 判断任务类型，选择检查子集
根据任务类型确定适用的检查项（参见 references/task-checklist.md）：
| 任务类型 | 适用检查项 | 额外检查 |
|---|---|---|
| 测试任务 | #2 命名 + #5 前端约定 | AC 标注格式（覆盖 AC: AC-NN） |
| 实现任务 | 完整 #1~#7 | 配对测试任务已完成（tasks.md 中已打勾）；design.md 同步检查 |
| 基础设施任务 | #1 架构分层 + #4 数据库约定 + #6 信息泄漏 + #7 设计同步 | 无 |
| Worker 任务 | #1 架构 + #4 数据库 + #6 信息泄漏 + #7 设计同步 | tenant_id 通过 Celery headers 传递 |
任务类型判断规则：
- 文件路径包含
test/或__tests__/→ 测试任务 - 文件路径包含
domain/models/或common/errcode/或任务描述含"ORM""迁移""错误码""配置" → 基础设施任务 - 文件路径包含
worker/或任务描述含"Celery""异步任务" → Worker 任务 - 其他 → 实现任务
- 若任务同时包含测试和实现文件，按实现任务处理
Step 3: 按检查清单执行检查
逐项执行 references/task-checklist.md 中适用的检查项。
Step 4: 元数据交叉验证
- 文件范围：任务声明的目标文件是否实际存在，是否存在范围蔓延（修改了任务未声明的文件）
- 配对测试：若为实现任务，检查 tasks.md 中配对的测试任务是否已打勾 ✅
- 前置依赖：检查任务声明的依赖项是否已完成（tasks.md 中已打勾）
Step 5: 输出报告
按以下格式输出：
## Task Review: <task_id>
**任务**: <任务标题>
**类型**: 测试 / 实现 / 基础设施 / Worker
**文件**: <文件列表>
| # | 检查项 | 结果 | 说明 |
|---|--------|------|------|
| 1 | 架构分层 | PASS / FAIL / N/A | <若 FAIL，具体描述> |
| 2 | 命名规范 | PASS / FAIL / N/A | |
| 3 | 序列化约定 | PASS / FAIL / N/A | |
| 4 | 数据库约定 | PASS / FAIL / N/A | |
| 5 | 前端约定 | PASS / FAIL / N/A | |
| 6 | 信息泄漏 | PASS / FAIL / N/A | |
| 7 | 设计同步 | PASS / FAIL / N/A | <若 FAIL，列出 design.md 哪节应更新> |
**元数据验证**: 文件范围 PASS/FAIL | 配对测试 PASS/FAIL/N/A | 依赖 PASS/FAIL
**结果**: PASS / PASS_WITH_NOTES / NEEDS_FIXStep 6: 处理结果
| 结果 | 条件 | 动作 |
|---|---|---|
| PASS | 全部通过 | 告知用户可以打勾 |
| PASS_WITH_NOTES | 仅 MEDIUM 级提醒，无 HIGH | 告知用户可以打勾，列出提醒供参考 |
| NEEDS_FIX | 任何 HIGH 违规 | 列出需要修复的具体问题，修复后可再次调用 /task-review 重审（最多 1 轮重审） |
错误处理
feature_dir不存在 → 报告路径错误，停止tasks.md不存在 → 报告"找不到 tasks.md"，停止task_id不匹配 → 报告"未找到任务 <task_id>"，停止- 目标文件不存在 → 标记为 WARNING（文件可能尚未创建），继续检查其他文件
© dataelement, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
SKILL.md and 1 other file (references) in .claude/skills/task-review of dataelement/bisheng.
- SKILL.md
- references/task-checklist.md
Open the folder on GitHubat commit ed3b461
Compare with similar skills
Task-Level Code Review next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Task-Level Code Review this skilldataelement/bisheng | 12k | — | ~652 | Automated safety check: Pass | Apache-2.0 | |
| Requesting Code ReviewHezaoHezao/poirot | 250 | 5 repos | ~1.6k | Automated safety check: Pass | MIT | |
| Compound Engineering Code ReviewEveryInc/compound-engineering-plugin | 25k | — | ~2k | Automated safety check: Pass | MIT | |
| Tbdjlevy/strif | 131 | — | ~3.5k | Automated safety check: Pass | MIT | |
| Pre-PR Reviewyuga-hashimoto/and-code | 123 | — | ~710 | Automated safety check: Pass | MIT | |
| Code Reviewmodimihir07/agentic-os | 192 | — | ~222 | Automated safety check: Pass | MIT |
Similar skills
Compound Engineering Code Review
EveryInc/compound-engineering-plugin
Runs a staged pull request or diff review using selected reviewer personas, checking the change against its stated intent and project standards before producing findings.
Tbd
jlevy/strif
Git-native issue tracking (beads), coding guidelines, knowledge injection, and spec-driven planning for AI agents.
Pre-PR Review
yuga-hashimoto/and-code
Runs local checks and a repo-reviewer subagent over the whole branch diff before a pull request is opened, then records the approval in the PR description.
Code Review
modimihir07/agentic-os
Automated code review with checklist and quality gates. An agent skill from modimihir07/agentic-os.
Gstack Pre-Landing Review
LING71671/Open-ClaudeCode
Runs a staff-engineer style review over the current diff before merging, listing findings by severity with file, impact and a suggested fix.
More from dataelement/bisheng
All 10 skills in this repoBiSheng DOCX Builder
dataelement/bisheng
Builds or edits Word .docx documents inside BiSheng's code executor with python-docx, handling Chinese fonts, tables of contents, page numbers and official-document layout.
BiSheng PPTX Builder
dataelement/bisheng
Builds PowerPoint decks in BiSheng's code executor with python-pptx, from scratch, from a supplied template or by rewriting an existing file, with Chinese layout rules.
BiSheng XLSX Workbook Builder
dataelement/bisheng
Builds, edits and cleans Excel workbooks inside BiSheng's code executor using openpyxl, with LibreOffice recalculation and a pre-delivery check.
BiSheng SDD Document Review
dataelement/bisheng
Reviews BiSheng spec, design and tasks documents with checklists for PRD gaps, handover readiness and acceptance traceability, producing a report or an LGTM.
BiSheng Approval Module Reference
dataelement/bisheng
Maps BiSheng's approval-center architecture to exact service files and methods, so a change to approval logic can be located without searching the whole repo.
Feature-Level Code Review
dataelement/bisheng
Runs a seven-dimension code review on a finished feature, comparing the diff against a base branch and the feature's spec, design and task files.
Related
Categories
Questions about Task-Level Code Review
What does Task-Level Code Review do?
Runs a light convention check on one finished spec-driven task, choosing checks by task type and ending in pass, pass-with-notes or needs-fix. Written in Chinese, this is the first level of a two-level review.md` is done, you run `/task-review` with the feature directory and task ID.
When should I use Task-Level Code Review?
Task-Level Code Review fits situations like: checking a just-finished spec-driven task before ticking it off; catching architecture-layer or naming violations early, at task level; re-reviewing a task after fixing high-severity findings.
How do I install Task-Level Code Review in Claude Code?
Run `npx skills add dataelement/bisheng --skill task-review -a claude-code`. Or copy the skill folder (.claude/skills/task-review in dataelement/bisheng) into .claude/skills/task-review in your project. Claude Code loads it when a task matches its description.
How do I install Task-Level Code Review in Codex?
Run `npx skills add dataelement/bisheng --skill task-review -a codex`. Or copy the skill folder (.claude/skills/task-review in dataelement/bisheng) into .agents/skills/task-review in your project. Codex loads it when a task matches its description.
Can I use Task-Level Code Review in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add dataelement/bisheng --skill task-review -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/task-review, .gemini/skills/task-review, .github/skills/task-review and .opencode/skills/task-review in your project.
What does Task-Level Code Review need to run?
SKILL.md names no scripts, command-line tools or credentials: Task-Level Code Review is instructions for the agent only. Our summary lists: A feature directory that contains a `tasks.md` file.
Does Task-Level Code Review access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Task-Level Code Review safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Task-Level Code Review use?
Task-Level Code Review is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Task-Level Code Review use?
About 652 tokens (SKILL.md is roughly 2.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.1k tokens, read only when the agent opens those files.
What are the alternatives to Task-Level Code Review?
Skills that share tags, products or a category with Task-Level Code Review: Requesting Code Review (HezaoHezao/poirot, 250 stars), Compound Engineering Code Review (EveryInc/compound-engineering-plugin, 25k stars), Tbd (jlevy/strif, 131 stars) and Pre-PR Review (yuga-hashimoto/and-code, 123 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Task-Level Code Review?
dataelement (a GitHub organization) maintains it in dataelement/bisheng, which has 12,028 GitHub stars. The repository holds 10 skills in this directory. The repository was last updated on October 8, 2026.
Source: dataelement/bisheng on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.