Individual skill
team
Invite humans into a workspace, propose collaborators for a project, and inspect the current human team. Use when the user says things like "let's bring X into this", "who else is on this", "add Bob to Q2 launch", or asks for a team roster. NEVER mints invites silently — the human must press a button.
sdk/cloud/catalog/skills/team/SKILL.md
Source commit f83b859a173802a73d978197e92328ab2bf57a6e
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- No
- Repository scan complete
- No
- Skill scan complete
- Yes
- Observed behavior
- Not checked yet
- Last checked
- Nov 3, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.
Repository context
- Repository stars (not skill ratings)
- 1,364
- Repository last updated
- Oct 22, 2026
- License
- MIT
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- agent-communication Repository topic
- agent-memory Repository topic
- ai-agents Repository topic
- claude Repository topic
- cursor Repository topic
- docker Repository topic
- knowledge-base Repository topic
- llm-tools Repository topic
- mcp Repository topic
- mcp-server Repository topic
- engineering Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
SKILL.mdcore · 5,862 bytes
Read excerpt
# Team
Use this skill when work needs another human in the loop. The workspace + project surfaces collectively decide *who can see and act on* tasks, conversations, and assets in a workspace; this skill is how the agent exposes those surfaces to the user.
## ⛔ Hard rules — human-in-loop
1. **Never auto-send an invite.** Agents may propose who should be invited and why. This skill's supported route for the final action is the human click in workspace settings. Even when the user asks to send an invite, route them to that UI; do not invent an invite CLI or bypass the supported confirmation path with a raw endpoint.
2. **Never expose member emails.** The `/api/im/workspaces/:id/members` response carries imUserIds, not addresses. If the user asks "what's Bob's email", say you can see Bob's username + display name but not their email.
3. **Distinguish workspace member from project member.** Workspace member = ACL boundary across the whole workspace (chat, tasks, assets). Project member = a sub-scope inside that workspace with its own role enum (owner/contributor/observer). Adding someone to a project requires them to be a workspace member first — surface that constraint when proposing.
4. **Never silently re-add a removed member.** If a user was removed from the workspace, their project memberships were cascade-deleted on purpose (release201/16 §3.2.3). Treat re-adding as a new collaboration event that needs explicit user consent, not a "fix it" automation.
## When to use
- The user names someone who isn't already on the workspace: *"Have Maya look at this"*, *"Loop Bob in on Q2 launch"*.