Driving Seat
Keeps the developer in control of every decision that matters while building or changing software. Turns vague goals into measurable intent, maps decisions up front, asks before any significant choice the spec doesn't cover, logs every choice made without asking, and checks that the developer understands decisions instead of delegating them. Use this skill whenever the user says "keep me in the driving seat", "ask before you decide", "don't make decisions for me", "I want to own the decisions", or invokes /driving-seat, and whenever they are building, changing, or fixing a production system an
Install
npx skills add Tanmay-Ts/driving-seat --skill driving-seatgit clone --depth 1 https://github.com/Tanmay-Ts/driving-seat.git /tmp/driving-seat
mkdir -p ~/.claude/skills
cp -r /tmp/driving-seat/. ~/.claude/skills/driving-seatPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# Driving Seat Production systems fail quietly when an assistant fills gaps on its own: a retry policy nobody chose, an error that gets swallowed, a dependency nobody approved. This skill exists so that every decision that matters has a human owner who understands it. Your role is not to do less. It's to make decisions *visible*, *recommended*, and *owned*: - Visible: no significant choice happens silently. - Recommended: every question comes with your opinion and the reason for it, so the developer never has to work it out alone. - Owned: the developer picks, and can say why. Asking about everything is not the goal. A gate that fires 200 times gets rubber-stamped (approved without being read), which is worse than no gate, because everyone believes a review happened. Ask about the decisions that matter, and make everything else auditable. ## Phase 1: Intent Ask up front only the questions whose answers change the plan: 1. **Intent:** what are we building or fixing, and why? 2. **Success metric:** how will we know it worked? 3. **Hard constraints:** stack, budget, deadline, compliance, existing systems it must fit into. 4. **Must never happen:** the failures that are unacceptable (data loss, downtime, leaked PII, wrong charges). Everything else gets asked just in time, when the decision actually comes up. Developers often can't answer detailed questions before seeing something concrete, and guessed answers look authoritative later. ### Keep the first message short
Details
- Skill name
driving-seat- Repository
- Tanmay-Ts/driving-seat
- License
- MIT
- Repo updated
- 20 October 2026
FAQ
What does the Driving Seat skill do?
Keeps the developer in control of every decision that matters while building or changing software. Turns vague goals into measurable intent, maps decisions up front, asks before any significant choice the spec doesn't cover, logs every choice made without asking, and checks that the developer understands decisions instead of delegating them. Use this skill whenever the user says "keep me in the driving seat", "ask before you decide", "don't make decisions for me", "I want to own the decisions", or invokes /driving-seat, and whenever they are building, changing, or fixing a production system an
How do I install the Driving Seat skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add Tanmay-Ts/driving-seat --skill driving-seat. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Driving Seat popular?
It lives in Tanmay-Ts/driving-seat, which has 10 GitHub stars and 1 forks.
Related skills
Saudi Regulatory Compliance
9ibnsawad/Saudi-Regulatory-Compliance-Claude-Skill
Senior cybersecurity, cloud, and data-protection compliance advisor for organisations in Saudi Arabia. Trigger for: NCA ECC-2:2024, CCC-2:2024, DCC-1:2022, CSCC-1:2019, OTCC-1:2022, TCC-1:2021, OSMACC-1:2021, CGEC-1:2019, CGESP-1:2019, NCS-1:2020; SAMA Cybersecurity/IT Governance/Business Continuity Frameworks; CST Cloud Computing Regulatory Framework, CST CRF RT08; PDPL and Implementing Regulations, Personal Data Transfer Regulation; NDMO Data Management Standards; SDAIA guidance including the AI Risk Management Framework (SDAIA-P145). Also trigger for gap assessments, audit readiness, breach
Codemap
10beadnall/codemap
Turns a codebase into an explorable isometric architecture diagram — a self-contained HTML artifact with blocks on a grid, a component index, a what-it-does/how-it's-built panel, and real data snippets travelling along the edges that the reader can stop and inspect. Use this whenever someone asks for a codemap by name — "create a codemap", "codemap this repo", "make a codemap of github.com/owner/name" — and equally when they want to see, map, diagram or visually explain how a system fits together: "diagram this codebase", "show me the architecture", "map how this works", "visualise the data fl
Compliance Audit
11oliwoodman/compliance-audit-skill
Audit a website, app, or product's codebase for GDPR and EU AI Act compliance, then hand back a ranked list of plain-English fixes. Asks the owner the handful of questions the code cannot answer, reads the actual code (forms, API routes, dependencies, env vars, database schema, policy pages) to work out what personal data the product really collects and who it is shared with, then checks that against what the law requires. It exists to catch the gaps a pasted-in policy template hides. MANDATORY TRIGGERS: 'compliance audit', 'run compliance-audit', 'audit my compliance', 'audit this for GDPR',
Ahrefs Domain Rating Free
9AIGC-Hackers/ahrefs-domain-rating-free-skill
Use when checking Ahrefs Domain Rating, DR, website authority, domain authority, backlink profile strength, or free Ahrefs public API data for a domain or URL without an API key.
Deconstruct Skill
11luoming-hu/deconstruct-skill
静态拆解本地文件、粘贴内容或代码仓库链接中的 Codex Skill,生成一份简洁、通俗、可下载的 Markdown 报告。报告只用六个章节直接说明最终交付、具体动作、步骤顺序、角色分工、规则所防的问题和可复用做法;不复制原文、不逐段批注、不列证据账本或源码引用,也不输出问题与改进建议。用户要求拆解、逆向分析、讲解或解释一个 Skill 为什么这样设计时使用。仅执行、安装或创建普通 Skill 时不要使用,除非用户同时要求先分析。
Opencode Automator
9alphaparkinc/genpark-skill-opencode-automator
OpenCode Automator — High-performance execution agent that utilizes dynamic code generation (Python/JS) to automate complex, multi-step workflows with mathematical precision and API reliability.
Browse the most popular Claude skills, all software development skills, or write your own with the skills generator.