Guide
Cursor team rules: how admins set rules for the whole team
Cursor Team Rules are instructions a Team or Enterprise admin writes once in the Cursor dashboard, and they apply to every member's Agent chats across all the team's repositories; an admin can enforce a rule so nobody can switch it off, or leave it optional. They're free-form text, can be limited to certain files with a glob pattern, and take precedence over project and user rules when guidance conflicts. Use team rules for standards that hold all the time and skills for methods the agent should load only when the task matches; skills kept in AI Skill Pod reach Cursor and everyone's Claude or ChatGPT (Pro, Business, Enterprise) from one invite link.
Short answer
- Where: the Cursor dashboard, by an admin.
- Who gets them: every member, in every repository.
- Enforced: can't be switched off.
- Precedence: team, then project, then user rules.
Cursor's four kinds of rules
| Kind | Where it lives | Applies to |
|---|---|---|
| Team Rules | The Cursor dashboard (Team and Enterprise) | Every member, across all the team's repositories |
| Project Rules | .cursor/rules in the repository | That codebase, for anyone who opens it |
| User Rules | One person's Cursor settings | That person's Agent chats |
| AGENTS.md | A markdown file in the project | That project, a simpler alternative to project rules |
From Cursor's Rules docs, read 5 November 2026, 23:08 JST. When guidance conflicts, team rules win over project rules, and project rules over user rules.
Set a team rule
- Open cursor.com/dashboard/team-content as an admin and add a rule.
- Write it as plain text, short and specific, such as "Use the shared logger; never log tokens or customer emails."
- Add a glob pattern if it applies only to some files, such as **/*.sql for migration rules.
- Choose Enable this rule immediately, or leave it as a draft, and tick Enforce this rule if members mustn't switch it off.
Cursor's docs add that AI guidance shouldn't be your only security control, even when a rule is enforced.
A worked example: three rules and one skill
Illustration: a made-up business. A 20-developer software company on Cursor Teams has style rules copied into a dozen repositories, each slightly different. The engineering manager:
- Creates three team rules: the logging rule (enforced), the testing rule (enforced) and a SQL migration rule limited to **/*.sql.
- Deletes the copies from the repositories' .cursor/rules, keeping only what's specific to each codebase.
- Moves the release-notes method out of the rules into a skill in the company's pod, so the agent loads it only when someone asks for release notes, and the product managers get it in Claude too.
AI Skill Pod is this site's product. Pod pricing: Free up to 3 members and 10 skills; Team $5 per seat a month from 3 seats.
Next step: the same skills in every coding tool
If the team also uses Claude Code or Codex, keep one copy of each skill for all of them. Sync skills to Claude Code, Codex, Cursor and OpenCode shows how.
Skills for this
- AI practiceTurn a prompt into a skillTurns a prompt that works into a skill the whole team can reuse, with a valid name, a when-to-use description, steps, an output template, an example, and checks.
- AI practiceContext brief for an AI assistantWrites instructions for an AI assistant or agent: role, audience, can and can't, sources of truth, tools, format, examples, edge cases, and handoff.
- ProductRelease notes from merged changesWrites customer-facing release notes from merged PRs or tickets, grouped New, Improved, Fixed, benefit-first, plus an internal changelog.
Questions
What are Cursor team rules?
Rules a Team or Enterprise admin writes in the Cursor dashboard that apply to every member. They're included in Agent chats across all the team's repositories, take precedence over project and user rules when guidance conflicts, and can be limited to certain files with a glob pattern.
Can members turn off a Cursor team rule?
Only if it isn't enforced. An admin can tick Enforce this rule, and then members can't disable it. Otherwise each member can switch it off under Team Rules in Customize.
Should a team method be a rule or a skill in Cursor?
A rule for standards that apply to all the work, such as code style or what never to log; a skill for a method the agent should load only when the task calls for it, such as writing release notes. Rules sit in every chat's context; skills load when relevant.
AI skill
Have your AI turn a prompt that works into a skill your team can reuse
Put this skill in AI Skill Pod and everyone on your team gets it in their Claude, Cursor or ChatGPT (Pro, Business, Enterprise): one invite link, free to start. Turn a prompt into a skill has your AI turn a long team rule that's really a method into a skill the agent loads only when the task matches: a valid name, a description that says when to use it, numbered steps, an output template and checks.
14 days free · No card needed
Sources
- Cursor Docs: Rules — read at 23:08 JST: Project Rules in .cursor/rules, User Rules, Team Rules and AGENTS.md; Team Rules on Team and Enterprise plans, managed from the dashboard, with Enable this rule immediately and Enforce this rule; non-enforced rules can be switched off under Team Rules in Customize; free-form text with optional glob patterns; included in Agent (Chat) across all the team's repositories; precedence Team, then Project, then User; AI guidance shouldn't be the only security control (read 2026-11-05)
- Cursor Docs: Agent Skills — read at 23:21 JST: Cursor discovers skills from skill directories, and the agent decides when they're relevant; also loads skills from .claude/skills and .codex/skills (read 2026-11-05)
- Cursor Docs: Team Pricing — read at 23:08 JST: Teams includes a team marketplace for internal rules, skills and plugins (read 2026-11-05)