IntuitivePhella/AI-PhellOS
Portable AI development workflow template and CLI for spec-driven, test-aware software delivery with Codex, Claude Code, and other coding agents.
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 IntuitivePhella/AI-PhellOS/.claude ./rig-ai-phellos # inspect, then merge into .claude/ 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.
$ curl -fsSL --create-dirs -o .claude/agents/architect.md https://raw.githubusercontent.com/IntuitivePhella/AI-PhellOS/main/.claude/agents/architect.md $ curl -fsSL --create-dirs -o .claude/agents/implementer.md https://raw.githubusercontent.com/IntuitivePhella/AI-PhellOS/main/.claude/agents/implementer.md $ curl -fsSL --create-dirs -o .claude/agents/orchestrator.md https://raw.githubusercontent.com/IntuitivePhella/AI-PhellOS/main/.claude/agents/orchestrator.md $ curl -fsSL --create-dirs -o .claude/agents/product.md https://raw.githubusercontent.com/IntuitivePhella/AI-PhellOS/main/.claude/agents/product.md $ curl -fsSL --create-dirs -o .claude/agents/qa.md https://raw.githubusercontent.com/IntuitivePhella/AI-PhellOS/main/.claude/agents/qa.md $ curl -fsSL --create-dirs -o .claude/agents/release.md https://raw.githubusercontent.com/IntuitivePhella/AI-PhellOS/main/.claude/agents/release.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/IntuitivePhella/AI-PhellOS/main/.claude/agents/reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/security.md https://raw.githubusercontent.com/IntuitivePhella/AI-PhellOS/main/.claude/agents/security.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"permissions": {
"deny": [
"Bash(rm -rf:*)",
"Bash(curl:*)",
"Bash(wget:*)",
"Bash(kubectl:*)",
"Bash(terraform apply:*)",
"Bash(aws:*)",
"Bash(gcloud:*)",
"Bash(az:*)",
"Bash(npm publish:*)",
"Bash(git push --force:*)"
]
}
} Subagents (8)
| architect model: inherit | Evaluate architecture, data model, APIs, dependencies, stack choices, integrations, and trade-offs. Use for new projects, multi-module features, framework adaptation, and risky technical decisions. |
| implementer model: inherit | Execute one ready story with tests first and the smallest useful code change. Use only after Definition of Ready is satisfied and scope, allowed files, forbidden areas, and verification commands are k |
| orchestrator model: inherit | Route AI development work through the smallest safe workflow, squad level, context packs, quality gates, brainstorming, and stop conditions. Use at the start of every non-trivial task and before escal |
| product model: inherit | Clarify product problem, target user, scope, non-goals, UX implications, brainstorming, and testable acceptance criteria. Use for vague ideas, new projects, new features, behavior changes, and ambiguo |
| qa model: inherit | Validate acceptance criteria, test strategy, edge cases, regression risk, and release confidence. Use after implementation, for bug reproduction, test planning, and autonomous phase validation. |
| release model: inherit | Assess release readiness, rollback, deployment risk, known issues, operational safety, and release notes. Use when user-visible behavior, migrations, deployment, or autonomous output acceptance is inv |
| reviewer model: inherit | Perform senior engineering review for maintainability, simplicity, architecture consistency, test quality, and regression risk. Use after code changes, refactors, broad diffs, or autonomous output. |
| security model: inherit | Review sensitive changes involving auth, authorization, billing, secrets, personal data, uploads, APIs, webhooks, integrations, infrastructure, migrations, and multi-tenant boundaries. |
Permissions
deny (10)
Bash(rm -rf:*)
Bash(curl:*)
Bash(wget:*)
Bash(kubectl:*)
Bash(terraform apply:*)
Bash(aws:*)
Bash(gcloud:*)
Bash(az:*)
Bash(npm publish:*)
Bash(git push --force:*)
ask (0)
—
allow (16)
Bash(git status:*)
Bash(git diff:*)
Bash(git log:*)
Bash(ls:*)
Bash(cat:*)
Bash(find:*)
Bash(grep:*)
Bash(rg:*)
Bash(npm test:*)
Bash(npm run test:*)
Bash(npm run build:*)
Bash(pnpm test:*)
Bash(pnpm build:*)
Bash(pytest:*)
Bash(ruff check:*)
Bash(mypy:*)
Similar rigs
Trolzie/claude-loop-systems-demo
A small demo of Markdown-as-system AI dev loops with Claude Code: issues → agents → skills → verification.
Pragmatist 480 tok ·
homeofe/improvements
Portable, LLM-agnostic framework for multi-model AI workflows: model routing, structured agent handoff (AAHP), and a phased agent pipeline. MIT.
Orchestrator 1.4k tok ·
xhulz/nina
Harness orchestration for Claude Code: composes a gated multi-agent pipeline into projects from versioned layers, holds its loop caps with hooks, measures it from transcripts, and turns recurring mistakes into rules.
Orchestrator 9.3k tok ·
Saba-Var/claude-ward
A read-only tripwire that watches Claude Code's local config and reports suspicious changes.
Pragmatist 1.8k tok ·