Scrollout/orbit
Orbit: a Claude Code plugin that gives one person a small software team. Role agents, a /decide PM, an operator running worker subagents in parallel, proof-checked TDD and a session ledger. A guard denies destructive git/GitHub actions, ask
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code $ npx degit Scrollout/orbit/agents ./rig-orbit/agents $ npx degit Scrollout/orbit/commands ./rig-orbit/commands $ npx degit Scrollout/orbit/skills ./rig-orbit/skills $ npx degit Scrollout/orbit/hooks ./rig-orbit/hooks
MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit Scrollout/orbit/skills/intent .claude/skills/intent $ npx degit Scrollout/orbit/skills/okf .claude/skills/okf $ npx degit Scrollout/orbit/skills/operator .claude/skills/operator $ npx degit Scrollout/orbit/skills/peer .claude/skills/peer
$ curl -fsSL --create-dirs -o .claude/agents/architect.md https://raw.githubusercontent.com/Scrollout/orbit/main/agents/architect.md $ curl -fsSL --create-dirs -o .claude/agents/coder.md https://raw.githubusercontent.com/Scrollout/orbit/main/agents/coder.md $ curl -fsSL --create-dirs -o .claude/agents/designer.md https://raw.githubusercontent.com/Scrollout/orbit/main/agents/designer.md $ curl -fsSL --create-dirs -o .claude/agents/explorer.md https://raw.githubusercontent.com/Scrollout/orbit/main/agents/explorer.md $ curl -fsSL --create-dirs -o .claude/agents/operator.md https://raw.githubusercontent.com/Scrollout/orbit/main/agents/operator.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/Scrollout/orbit/main/agents/reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/security.md https://raw.githubusercontent.com/Scrollout/orbit/main/agents/security.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 16.2k rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(./.env.*)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(git push --force:*)",
"Read(**/*.pem)",
"Read(.env)",
"Read(~/.aws/**)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Bash(rm -rf *)",
"Bash(sudo *)",
"Bash(sudo:*)",
"Bash(rm -rf /)",
"Bash(git push -f:*)",
"Read(.env.*)",
"Bash(git reset --hard*)",
"Bash(rm -rf /*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(gh pr merge *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(gh pr merge:*)",
"Bash(gh pr create *)",
"Bash(git reset --hard:*)",
"Bash(npm publish:*)",
"Bash(git rebase *)"
]
}
} Skills (4)
Subagents (7)
| architect model: fable | The tech lead in /decide (hyperscale-operator mindset): signs off architecture, decision-record, roadmap and trade-off questions. Use after discovery, proposals and design review, never to explore. Gi |
| coder model: sonnet | Implements a specified step. Code, tests, fixes, migrations, long drafts of docs or decision-record text from a decision already made. Use proactively when the task is implementation with a clear spec |
| designer model: sonnet | Plays one engineer or one reviewer in a /decide design review, from the single focus the PM names (for example root-fix, simplicity-first, concurrency, ops cost, devil's advocate). As engineer, writes |
| explorer model: sonnet | Sweeps the code and docs. Finds where something lives, how a pattern is used, what a large file or a set of files says. Use proactively for any task that needs many files read, before thinking about t |
| operator model: inherit | Main-session agent for the operator of a team of agents; start it with `claude --agent orbit:operator` in the project's main checkout. Not for use as a subagent (a subagent can't start workers). |
| reviewer model: opus | Reviews a built step's diff against its spec in a fresh context, before the PM accepts it; with the first line `Mode: fix-round`, checks one fix round. Give it the step brief (spec, conditions, proof |
| security model: fable | The security officer in /decide: judges the proposals as an attacker, after design review and before the tech lead (first for a security finding or incident). Use on the full review and incident track |
Slash commands (3)
/decide/okf-init/unattended
Similar rigs
AegeanWorks/AIKit
Crash-resilient governance and delivery gates for AI coding agents: versioned task ledger, commit-level verification, secret scanning and per-project gates. Free for personal use under AL-1.1; commercial use under ACL-1.0.
Orchestrator 4.2k tok · · Q23
bpstack/harness
A reusable foundation for AI projects: rules, agents, templates, and commands.
Pragmatist 3.5k tok · · Q5
andreiciobotar/isshoni
A self-improving software factory for Claude Code, with a live terminal dashboard. Looks delightful.
Pragmatist 893 tok · · Q0
jmigual/dotfiles
Personal dotfiles
Pragmatist 479 tok · · Q0