teammate-protocol
Skill1 starsChanged 23 days agoMITView on GitHub
The rules that apply when a specialist is spawned as part of a pre-assembled Claude Code Agent Team rather than as a one-shot sub-agent. Preloaded into the five delivery-team specialists.
In this repository18 other files it gives its agents.
- AGENTS.mdAGENTS.md
- context7-cliSkill
- decision-protocolSkill
- frontend-designSkill
- gitlab-accessSkill
- hallmarkSkill
- harness-cleanerSkill
- image-genSkill
- k8s-debugSkill
- retroSkill
- sdd-workflowSkill
- ste100Skill
- team-wikiSkill
- technical-seoSkill
- ux-copySkill
- vector-genSkill
- web-design-guidelinesSkill
- writing-for-agentsSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 8
When you are spawned with a team_name — part of a pre-assembled Agent Team rather than a one-shot sub-agent — five protocols apply on top …
- Distinctive: line 13
Idling between turns is normal: Do not poll, and do not read idleness as the team being finished.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
When operating as a teammate on an Agent Team
When you are spawned with a team_name — part of a pre-assembled Agent Team rather than a one-shot sub-agent — five protocols apply on top of your role:
- Read
TaskListbefore anything else: what exists, what is assigned to you, what is blocked. Never create a task without first verifying no equivalent one exists — duplicates fragment ownership and waste the coordinator's triage. - Log every substantive peer consultation to
${CLAUDE_PLUGIN_ROOT}/.claude/collaboration-traces/<run-dir>/<your-role>-consultation-<topic>.md(who asked what, the response, how it resolved), then send the coordinator a one-line acknowledgement. Trivial pings need no log. - Route structural blockers through the coordinator first. Go peer-to-peer only when you already know whose answer it is, and copy the coordinator with a one-line summary of the substance.
- Idling between turns is normal. Do not poll, and do not read idleness as the team being finished. Wake on the next message.
Full protocol with the reasoning behind each rule: ${CLAUDE_PLUGIN_ROOT}/.claude/process/teammate-protocol.md.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

