Agent skill

BiSheng SDD Document Review

by dataelement in 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.

Apache-2.0Auto-check passedDevelopment

SKILL.md written in Chinese; this summary is our English description.

Install BiSheng SDD Document Review

skills CLI
$ npx skills add dataelement/bisheng --skill sdd-review -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install dataelement/bisheng sdd-review --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/dataelement/bisheng.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/sdd-review .claude/skills/sdd-review && rm -rf skills-src

Use ~/.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/

Facts

Skill name
sdd-review
GitHub stars
12k
Token cost
~717 tokens
SKILL.md length
167 words
Files
4 (incl. references)
Skills in repo
10
Repo updated
First seen
Licence
Apache-2.0

At a glance

Reviews BiSheng spec, design and tasks documents with checklists for PRD gaps, handover readiness and acceptance traceability, producing a report or an LGTM.

  • After writing spec.md, to check it against the PRD and architecture
  • 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
  • After writing design.md, to test whether a new agent could take over

What it does

Invoked as `/sdd-review` with a feature directory and a document type, it has three modes. Spec mode checks `spec.md` against the linked PRD, release contract and architecture docs with 14 checks and reports missing, format, conflict and other issues. Design mode applies 24 checks to `design.md`, built around a handover test: current state, decisions, pitfalls and contracts. Both wait for your confirmation before marking the document as reviewed in the status table of `tasks.md`.

Tasks mode runs automatically with 21 checks covering acceptance-criteria traceability, task breakdown quality and technical debt prevention. It prints issues with a severity, or LGTM; high and medium issues are fixed and reviewed again for up to two rounds, low ones are only recorded, and persistent problems are reported to you. Missing directories, documents or invalid types stop the review with a clear error. The text is in Chinese.

When your agent uses it

  • After writing spec.md, to check it against the PRD and architecture
  • After writing design.md, to test whether a new agent could take over
  • After writing tasks.md, to verify acceptance traceability and task quality

Example prompts

  • “/sdd-review features/v2.5.0/004-rebac-core spec”
  • “Review the design.md for the multi-tenant feature and list any pitfalls missing from it.”
  • “Run the tasks review on features/v2.5.0/001-multi-tenant and fix any high-severity issues.”

Requirements

  • A BiSheng-style feature folder with spec.md, design.md or tasks.md
  • Referenced docs such as the release contract and architecture notes

What it can do on your machine

Read from SKILL.md and the folder at commit cb9b77a. 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.

    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

BiSheng SDD Document Review loads about 717 tokens when it runs, and up to ~3.4k if it reads all its reference files. Until then it costs about 95 tokens; SKILL.md has 167 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~95
When it runs · the whole SKILL.md, loaded when a task matches
~717
With references · SKILL.md plus every file in references/, read only if the agent opens them
~3.4k

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 cb9b77a, republished under its Apache-2.0 licence (© dataelement). 167 words, ~717 tokens.

Download SKILL.mdSave it as .claude/skills/sdd-review/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
sdd-review
description
对 BiSheng 项目的 SDD 文档执行审查。 - spec:写完 spec.md 后调用,同时检查 PRD gap 和架构合规性,生成报告供用户参考 - design:写完 design.md 后调用,检查"接手测试"四要素(现状/决策/坑/契约),生成报告供用户参考 - tasks:写完 tasks.md 后自动调用,检查 AC 追溯、任务拆解质量和技术债预防 用法:/sdd-review <feature_dir> <doc_type>,doc_type 为 spec / design / tasks。 TRIGGER when: 用户完成了 SDD 的 spec.md / design.md / tasks.md 编写,或者用户使用 /sdd-review 命令,或者 Claude 完成了这些文件的编写后需要审查。

SDD Review Skill

调用方式

/sdd-review <feature_dir> <doc_type>

例:

/sdd-review features/v2.5.0/004-rebac-core spec
/sdd-review features/v2.5.0/001-multi-tenant tasks

审查流程

第一步:解析参数

从用户输入或调用上下文中提取:

  • feature_dir:特性目录路径(如 features/v2.5.0/004-rebac-core)
  • doc_type:文档类型,必须是 spec / design / tasks

若参数缺失或无效,向用户报告错误后停止。


spec 模式(辅助审查,不自动推进)

spec.md 合并了需求规范和技术设计,因此 spec 审查同时覆盖需求覆盖和架构合规检查。

第二步(spec):执行合并审查

读取文件:

  • <feature_dir>/spec.md(已写的规格文档)
  • spec.md 中"关联 PRD"字段指向的文件(若未标注,读取 docs/PRD/ 下与特性名最相关的文件)
  • features/v2.5.0/release-contract.md(不变量约束,确认 spec 未越界)
  • docs/architecture/02-backend-modules.md(后端模块架构)
  • docs/architecture/10-permission-rbac.md(权限体系)

按 references/spec-checklist.md 中的检查清单执行 14 项检查。

第三步(spec):展示报告,等待用户确认

向用户展示分析结果:

  • 无 gap / 无问题:告知"审查通过,可继续确认"
  • 有 gap / 有问题:展示每个问题(MISSING / FORMAT / CONFLICT / ISSUE),供用户决定是否修改

等待用户确认(唯一手动暂停点)。用户确认后,将 <feature_dir>/tasks.md 状态表中 spec.md 行更新为 ✅ 已评审。


design 模式(辅助审查,不自动推进)

design.md 是"现状快照 + 关键决策"文档,决定新 agent 接手时能否在不读代码情况下快速建立认知。审查重点是接手测试四要素:现状、决策、坑、契约。

第二步(design):执行审查

读取文件:

  • <feature_dir>/design.md(已写的设计文档)
  • <feature_dir>/spec.md(校验 design 没偏离需求)
  • <feature_dir>/tasks.md(若存在;校验偏差记录已回写)
  • features/_templates/design.md(模板结构基线)
  • features/v{X.Y.Z}/release-contract.md(关键约束、不变量)

按 references/design-checklist.md 中的检查清单执行 24 项检查。

第三步(design):展示报告,等待用户确认

向用户展示分析结果:

  • 无问题:告知"审查通过,可继续确认"
  • 有问题:展示每个问题(ISSUE 含 SEVERITY 和 LOC),供用户决定是否修改

等待用户确认(唯一手动暂停点)。用户确认后,将 <feature_dir>/tasks.md 状态表中 design.md 行更新为 ✅ 已评审(若状态表尚无该行,提示用户补一行 | design.md | 🔲 草稿 | ... |)。


tasks 模式(自动审查)

第二步(tasks):执行审查

读取文件:

  • <feature_dir>/tasks.md
  • <feature_dir>/spec.md(验收标准 + 技术方案)
  • features/v2.5.0/release-contract.md(领域归属 + 不变量)

按 references/tasks-checklist.md 中的检查清单执行 21 项检查。

第三步(tasks):处理审查结果

输出格式:

  • 有问题:ISSUE: <描述> | SEVERITY: high/medium/low | TASK: <T-NN 若适用>
  • 无问题:LGTM

处理逻辑:

  • LGTM → 更新 <feature_dir>/tasks.md 状态表,将 tasks.md 行改为 ✅ 已拆解
  • 有 high/medium ISSUE → 修复后重新审查(最多 2 轮)
  • low ISSUE → 记录但跳过
  • 2 轮后仍有 high/medium → 停止,向用户报告剩余问题

错误处理

  • feature_dir 不存在 → 报告路径错误,停止
  • doc_type 不是 spec / design / tasks → 报告参数错误,停止
  • spec.md 不存在 → 报告"找不到 spec.md,请先完成 spec",停止
  • design.md 不存在(design 模式)→ 报告"找不到 design.md,请先按 features/_templates/design.md 完成 design",停止
  • tasks.md 不存在(tasks 模式)→ 报告"找不到 tasks.md,请先完成 tasks",停止

© 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 3 other files (references) in .claude/skills/sdd-review of dataelement/bisheng.

  • SKILL.md
  • references/design-checklist.md
  • references/spec-checklist.md
  • references/tasks-checklist.md

Open the folder on GitHubat commit cb9b77a

Compare with similar skills

BiSheng SDD Document 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.

BiSheng SDD Document Review compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
BiSheng SDD Document Review this skilldataelement/bisheng12k—~717Automated safety check: PassApache-2.0
MVP Technical DesignKhazP/vibe-coding-prompt-template3.1k—~512Automated safety check: PassMIT
Module Completeness Checkfengshao1227/ccg-workflow5.9k—~473Automated safety check: NotesMIT
GAIA Agent Eval Scorecardamd/gaia1.6k—~2.6kAutomated safety check: PassMIT
Caveman Spec CompressionJuliusBrussee/cavekit1.1k—~721Automated safety check: PassMIT
Maintainer Docs Legibility Checkliaohch3/claude-tap3.3k—~561Automated safety check: PassMIT

Similar skills

  • MVP Technical Design

    KhazP/vibe-coding-prompt-template

    Writes an MVP technical design from agreed requirements, covering architecture, data ownership, integration contracts, deployment and tradeoffs, then hands off to the next stage.

    3.1k GitHub stars~512 tokensUpdated 3 days ago
    DevelopmentAuto-check passed
  • Module Completeness Check

    fengshao1227/ccg-workflow

    Scans a module directory for the required README.md and DESIGN.md plus recommended files and reports what is missing, so a module is not delivered incomplete.

    5.9k GitHub stars~473 tokensUpdated 22 days ago
    DevelopmentAuto-check: notes
  • Adds a release eval scorecard to a GAIA hub agent by writing a harness adapter, running a real eval, and wiring the result into the agent's README and release gate.

    1.6k GitHub stars~2.6k tokensUpdated yesterday
    DevelopmentAuto-check passed
  • Caveman Spec Compression

    JuliusBrussee/cavekit

    Compresses SPEC.md writes and spec-referencing prose into terse, symbol-heavy fragments that drop articles, filler and hedging while keeping facts intact.

    1.1k GitHub stars~721 tokensUpdated 1 mo ago
    DevelopmentAuto-check passed
  • Runs local checks on maintainer docs for stale review dates, missing manifest paths and plan state drift, matching what the CI workflow enforces.

    3.3k GitHub stars~561 tokensUpdated 15 days ago
    DevelopmentAuto-check passed
  • Aholo Renderer Package Rules

    manycoretech/aholo-viewer

    Work on the @manycore/aholo-viewer package. Use only when the user explicitly asks for renderer source, package build, public exports, JSDoc/API docs…

    1.1k GitHub stars~191 tokensUpdated 13 days ago
    DevelopmentAuto-check passed

More from dataelement/bisheng

All 10 skills in this repo
  • BiSheng 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.

    12k GitHub stars~2.6k tokensUpdated 7 days ago
    Auto-check passed
  • 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.

    12k GitHub stars~1.8k tokensUpdated 7 days ago
    Auto-check passed
  • Builds, edits and cleans Excel workbooks inside BiSheng's code executor using openpyxl, with LibreOffice recalculation and a pre-delivery check.

    12k GitHub stars~1.7k tokensUpdated 7 days ago
    Auto-check passed
  • Task-Level Code Review

    dataelement/bisheng

    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.

    12k GitHub stars~652 tokensUpdated 7 days ago
    Auto-check passed
  • 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.

    12k GitHub stars~4k tokensUpdated 7 days ago
    Auto-check passed
  • 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.

    12k GitHub stars~1k tokensUpdated 7 days ago
    Auto-check passed

Questions about BiSheng SDD Document Review

What does BiSheng SDD Document Review do?

Reviews BiSheng spec, design and tasks documents with checklists for PRD gaps, handover readiness and acceptance traceability, producing a report or an LGTM. Invoked as `/sdd-review` with a feature directory and a document type, it has three modes.md` against the linked PRD, release contract and architecture docs with 14 checks and reports missing, format, conflict and other issues.

When should I use BiSheng SDD Document Review?

BiSheng SDD Document Review fits situations like: after writing spec.md, to check it against the PRD and architecture; after writing design.md, to test whether a new agent could take over; after writing tasks.md, to verify acceptance traceability and task quality.

How do I install BiSheng SDD Document Review in Claude Code?

Run `npx skills add dataelement/bisheng --skill sdd-review -a claude-code`. Or copy the skill folder (.claude/skills/sdd-review in dataelement/bisheng) into .claude/skills/sdd-review in your project. Claude Code loads it when a task matches its description.

How do I install BiSheng SDD Document Review in Codex?

Run `npx skills add dataelement/bisheng --skill sdd-review -a codex`. Or copy the skill folder (.claude/skills/sdd-review in dataelement/bisheng) into .agents/skills/sdd-review in your project. Codex loads it when a task matches its description.

Can I use BiSheng SDD Document 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 sdd-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/sdd-review, .gemini/skills/sdd-review, .github/skills/sdd-review and .opencode/skills/sdd-review in your project.

What does BiSheng SDD Document Review need to run?

SKILL.md names no scripts, command-line tools or credentials: BiSheng SDD Document Review is instructions for the agent only. Our summary lists: A BiSheng-style feature folder with spec.md, design.md or tasks.md; Referenced docs such as the release contract and architecture notes.

Does BiSheng SDD Document 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 BiSheng SDD Document 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 BiSheng SDD Document Review use?

BiSheng SDD Document 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 BiSheng SDD Document Review use?

About 717 tokens (SKILL.md is roughly 2.9k 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 2.7k tokens, read only when the agent opens those files.

What are the alternatives to BiSheng SDD Document Review?

Skills that share tags, products or a category with BiSheng SDD Document Review: MVP Technical Design (KhazP/vibe-coding-prompt-template, 3.1k stars), Module Completeness Check (fengshao1227/ccg-workflow, 5.9k stars), GAIA Agent Eval Scorecard (amd/gaia, 1.6k stars) and Caveman Spec Compression (JuliusBrussee/cavekit, 1.1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains BiSheng SDD Document Review?

dataelement (a GitHub organization) maintains it in dataelement/bisheng, which has 12,027 GitHub stars. The repository holds 10 skills in this directory. The repository was last updated on September 30, 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.