opencode-team
AGENTS.md69 starsChanged 19 days agoMITView on GitHub
OpenCode Team defines 14 agent roles in agents/*.md. Every role uses mode: all, so it is reachable three ways: The orchestrator is the default agent. It routes an idea through the pipeline and asks for approval at each gate. A specialist owns one stage and writes that stage's artifact. In the build stage, fe and be write specialist reports and the orchestrator consolidates them into .team/07-build.md. Every agent is reachable directly: Directly invoked agents follow the same memory protocol as…
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,439 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository15 other files it gives its agents.
- team/AGENTS.mdAGENTS.md
- team-architectureSkill
- team-backlogSkill
- team-brandSkill
- team-buildSkill
- team-disciplineSkill
- team-docsSkill
- team-marketingSkill
- team-market-researchSkill
- team-memorySkill
- team-prdSkill
- team-qaSkill
- team-releaseSkill
- team-security-reviewSkill
- team-uxSkill
The file
We split this agents.md into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are 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.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 13
Every agent is reachable directly:
- Very specific: line 35
Create agents/<role>.md with frontmatter (description, mode: all, model (optional), permissions as an ordered array of {action, resource, …
- Very specific: line 37
Add a command in commands/<role>.md if you want a direct entry point.
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md 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
How agents work
OpenCode Team defines 14 agent roles in agents/*.md. Every role uses mode: all, so it is reachable three ways:
@agent-nameto mention one directly- Shift+Tab (or
/agents) to switch agents - the orchestrator can launch it as a subagent through the
subagenttool
The orchestrator is the default agent. It routes an idea through the pipeline and asks for approval at each gate. A specialist owns one stage and writes that stage's artifact. In the build stage, fe and be write specialist reports and the orchestrator consolidates them into .team/07-build.md.
Direct invocation
Every agent is reachable directly:
@analyst
@pm
@po
@ux
@brand
@architect
@fe
@be
@qa
@security
@docs
@devops
@marketingDirectly invoked agents follow the same memory protocol as subagents: read .team/memory/INDEX.md and their own log, then write at most 5 lines to their own log.
Adding a role
- Create
agents/<role>.mdwith frontmatter (description,mode: all,model(optional),permissionsas an ordered array of{action, resource, effect}). The filename becomes the agent id; do not addname. - Reference it in
team.yamlasagent: <role>. - Add a command in
commands/<role>.mdif you want a direct entry point.
Full list
Agent reference — generated from the filename and frontmatter: description, mode and model.
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.

