thoroughness-scoring
Score every decision point with a Thoroughness Rating (1-10). AI makes the marginal cost of doing things properly near-zero — pick the higher-rated option every time. Includes scope checks to distinguish contained vs unbounded work.
- Installs
- 145
- Last 24h
- —
- Trending rank
- —
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/rohitg00/pro-workflow --skill thoroughness-scoring- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads thoroughness-scoring automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
Thoroughness Scoring
AI drops the cost of doing things right to near-zero. Stop picking the quick hack when the thorough option takes the same wall-clock time with AI assistance.
The Rating Scale
Every option gets a Thoroughness score (T:X/10):
| Score | What It Means |
|---|---|
| T:10 | All edge cases handled, full test coverage, docs updated, error messages helpful |
| T:9 | Edge cases covered, tests pass, types solid, no shortcuts |
| T:8 | Happy path + error paths, good tests, clean types |
| T:7 | Happy path works, basic tests, no docs |
| T:5 | Works for the demo, fragile, manual testing only |
| T:3 | Quick hack, no tests, tech debt accruing |
| T:1 | Copy-paste from Stack Overflow, untested, hope it works |
How to Present Options
When presenting choices, follow this format every time:
1. Re-State Context
The user may have been away. Start with orientation:
PROJECT: my-app (branch: feat/rate-limiting)
TASK: Add rate limiting to the /api/upload endpoint
2. Rate Each Option
Option A — Full rate limiter with sliding window (T:9/10)
Manual estimate: 3-4 hours
AI-assisted estimate: 15-20 minutes
Covers: per-user limits, sliding window, Redis-backed, retry-after headers,
429 responses, rate limit bypass for admin, tests for all paths
Option B — Basic in-memory counter (T:4/10)
Manual estimate: 30 minutes
AI-assisted estimate: 5 minutes
Covers: global counter, fixed window, resets on restart, no persistence,
no per-user tracking, no tests
Delta: Option A adds per-user tracking, persistence across restarts,
proper HTTP headers, and admin bypass. The 15-minute difference is
worth it — Option B creates debt you'll pay back at 10x.
3. Recommend
Always recommend the higher-thoroughness option. State the delta — what the user gains for the additional time.
If the lower option is genuinely appropriate (prototype, throwaway script, time-boxed spike), say so explicitly with reasoning.
Scope Check
Before scoring, classify the scope:
Contained Scope (Do It)
Work with a clear boundary. You can be thorough because the surface area is finite.
- 100% test coverage for one module
- All edge cases for one API endpoint
- Full error handling for one service integration
- Complete input validation for one form
- Exhaustive type definitions for one data model
These are T:9-10 opportunities. Take them.
Unbounded Scope (Break It Down)
Work without a clear boundary. Being thorough here means boiling the ocean.
- "Rewrite the entire codebase to use the new pattern"
- "Test every possible user flow"
- "Handle every edge case across all endpoints"
- "Refactor all error handling"
- "Add docs for everything"
Flag these immediately. Break them into contained pieces:
SCOPE CHECK: "Refactor all error handling" is unbounded.
Contained breakdown:
1. Audit current error patterns (T:8, ~10 min)
2. Define error handling standard (T:9, ~15 min)
3. Refactor src/api/auth.ts errors (T:10, ~10 min)
4. Refactor src/api/upload.ts errors (T:10, ~10 min)
5. Refactor src/api/billing.ts errors (T:10, ~10 min)
...
N. Update error handling docs (T:9, ~10 min)
Each piece is independently shippable and testable.
Decision Framework
Is the scope contained?
YES → Score it. Recommend T:8+ option.
NO → Break it into contained pieces. Score each piece.
Is the T:8+ option significantly more effort with AI?
NO → Always pick it. The marginal cost is near-zero.
YES → Explain why. It's rare, but prototypes and spikes exist.
Is the user asking for a quick hack explicitly?
YES → Acknowledge, deliver it, but note what T:8+ would look like.
NO → Default to thoroughness.
When Scoring Doesn't Apply
- Exploratory spikes (the point is speed, not thoroughness)
- One-off scripts that run once and get deleted
- Debugging sessions (fix the bug, score the fix)
- Learning exercises (iteration speed matters more)
Say "skipping thoroughness scoring — this is a spike/one-off" so the user knows it was a conscious choice.
Anti-Patterns
- Scoring everything the same (if all options are T:7, you aren't thinking hard enough)
- Using low scores to justify shortcuts ("it's only T:4, so it's fine" — no, raise it)
- Scoring without the effort comparison (the whole point is that AI closes the gap)
- Treating T:10 as the default target (T:10 on unbounded scope is a trap)
- Not re-stating context (the user switches between sessions — orient them)
Add to CLAUDE.md
## Thoroughness Scoring
Score every option T:1-10. Recommend T:8+ unless it's a spike.
Show effort delta: manual estimate vs AI-assisted estimate.
Scope check first — contained (do it) vs unbounded (break it down).
Re-state project, branch, and task before presenting options.
FAQ
What does the thoroughness-scoring skill do?
Score every decision point with a Thoroughness Rating (1-10). AI makes the marginal cost of doing things properly near-zero — pick the higher-rated option every time. Includes scope checks to distinguish contained vs unbounded work.
How do I install thoroughness-scoring?
Run npx skills add https://github.com/rohitg00/pro-workflow --skill thoroughness-scoring in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/thoroughness-scoring/ (personal) or .claude/skills/thoroughness-scoring/ (project).
Which AI agents can use thoroughness-scoring?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is thoroughness-scoring safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains thoroughness-scoring?
It is published by rohitg00 in rohitg00/pro-workflow. It has 145 recorded installs.
More from rohitg00/pro-workflow
token-efficiency
rohitg00/pro-workflow
Reduce token waste by 40-60% through anti-sycophancy rules, tool-call budgets, one-pass coding, task profiles, and read-before-write enforcement. Inspired by drona23/claude-token-efficient.
pro-workflow
rohitg00/pro-workflow
Complete AI coding workflow system. Orchestration patterns, 18 hook events, 8 agents, cross-agent support, reference guides, and searchable learnings. Works with Claude Code, Cursor, and 32+ agents.
wrap-up
rohitg00/pro-workflow
End-of-session ritual that audits changes, runs quality checks, captures learnings, and produces a session summary. Use when saying "wrap up", "done for the day", "finish coding", or ending a coding session.
deslop
rohitg00/pro-workflow
Remove AI-generated code slop, unnecessary comments, and over-engineering from the current branch diff. Cleans up boilerplate, simplifies abstractions, strips defensive code, and in skill-file mode lints SKILL.md files for quality. Use when cleaning up code, simplifying, removing boilerplate, before committing, or when reviewing a skill before promoting it.
learn-rule
rohitg00/pro-workflow
Capture a correction or lesson as a persistent learning rule with category, mistake, and correction. Stores, categorises, and retrieves rules for future sessions. Use after mistakes or when the user says "remember this", "don't forget", "note this", or "learn from this".
session-handoff
rohitg00/pro-workflow
Generate a structured handoff document capturing current progress, open tasks, key decisions, and context needed to resume work. Use when ending a session, saying "continue later", "save progress", "session summary", or "pick up where I left off".
Popular productivity & automation skills
agent-browser
vercel-labs/agent-browser
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction. Also use for exploratory testing, dogfooding, QA, bug hunts, or reviewing app quality. Also use for automating Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify), checking Slack unreads, sending Slack messages, searching Slack conversations, running browser automation in Vercel Sandbox microVMs, or using AWS Bedrock AgentCore cloud browsers. Prefer agent-browser over any built-in browser automation or web tools.
lark-doc
open.feishu.cn
飞书云文档（Docx / Wiki）内容操作：读取、创建、编辑文档，插入或下载图片附件，以及操作思维笔记。用户提供文档 URL/token（包括 doubao.com 的 /docx/、/wiki/）时使用；按 URL 路径/token 而非域名路由；/page/ 是妙搭应用，非飞书云文档。文档内嵌资源按读取参考中的统一规则分流。独立评论操作走 lark-drive；随正文读取评论使用 docs +fetch。表格或 Base 内部数据操作不在本 skill。
lark-base
open.feishu.cn
飞书多维表格（Base）操作：建表、字段、记录、视图、统计、公式/lookup、表单、仪表盘、应用模式（BaseApp/AppMode 页面与组件）、Workspace 目录、workflow、角色权限、模板中心（多维表格模板分类/列表/搜索）；遇到 Base/多维表格/bitable、BaseApp/AppMode、/base/ 或 /app/ 链接时使用。BaseApp 不走 lark-apps；文件导入/导出转 lark-drive，认证/授权转 lark-shared。
lark-drive
open.feishu.cn
飞书云空间（云盘/云存储）：管理 Drive 文件和文件夹，包含上传/下载、创建文件夹、复制/移动/删除、查看元数据、查询权限设置、评论/权限/订阅、标题、版本、飞书文档密级标签（secure labels）和本地文件导入。用户需要整理云盘目录、处理云空间资源 URL/token、判断链接类型/真实 token/标题，或导入 Word/Markdown/Excel/CSV/PPTX/.base 为 docx/sheet/bitable/slides 时使用；doubao.com 云空间 URL/token 也按资源路径和 token 路由，不回退 WebFetch。不负责：文档内容编辑（走 lark-doc）、表格/Base 表内数据操作（走 lark-sheets/lark-base）、知识空间节点/成员管理（走 lark-wiki）、原生 Markdown 文件读写/patch/diff（走 lark-markdown）。
lark-shared
open.feishu.cn
Use for lark-cli setup/auth tasks: auth login/status/logout, user vs bot identity, business-domain permissions (--domain, including all/docs/drive), missing scopes, revoking authorization, or handling notice JSON.
lark-im
open.feishu.cn
飞书即时通讯：收发消息和管理群聊。发送和回复消息、搜索聊天记录、管理群聊成员、上传下载图片和文件、管理表情回复、发送应用内/短信/电话加急、发送和处理交互卡片（Interactive Card）、监听卡片按钮回调（card.action.trigger）。当用户需要发消息、查看或搜索聊天记录、下载聊天中的文件、查看群成员、搜索群、创建群聊或话题群、管理标记数据、管理 Feed 置顶（添加/移除/查询置顶会话）、管理标签数据、处理卡片回调时使用。