tammai/ai-sdlc
Claude Code plugin for vibe coding on the AI-native SDLC playbook: intent → design → spec → plan → verify → review → ship. Runs on a Claude subscription, with opinionated stacks (Nuxt/Next + Cloudflare, Go + Postgres, Flutter, Tauri) and de
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 tammai/ai-sdlc/.claude ./rig-ai-sdlc # 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.
$ npx degit tammai/ai-sdlc/skills/architecture .claude/skills/architecture $ npx degit tammai/ai-sdlc/skills/fix .claude/skills/fix $ npx degit tammai/ai-sdlc/skills/intent .claude/skills/intent $ npx degit tammai/ai-sdlc/skills/learn .claude/skills/learn $ npx degit tammai/ai-sdlc/skills/plan .claude/skills/plan $ npx degit tammai/ai-sdlc/skills/policy .claude/skills/policy $ npx degit tammai/ai-sdlc/skills/review .claude/skills/review $ npx degit tammai/ai-sdlc/skills/setup .claude/skills/setup $ npx degit tammai/ai-sdlc/skills/ship .claude/skills/ship $ npx degit tammai/ai-sdlc/skills/spec .claude/skills/spec $ npx degit tammai/ai-sdlc/skills/stack .claude/skills/stack $ npx degit tammai/ai-sdlc/skills/status .claude/skills/status $ npx degit tammai/ai-sdlc/skills/system-design .claude/skills/system-design $ npx degit tammai/ai-sdlc/skills/triage .claude/skills/triage $ npx degit tammai/ai-sdlc/skills/uiux .claude/skills/uiux $ npx degit tammai/ai-sdlc/skills/vibe .claude/skills/vibe
$ curl -fsSL --create-dirs -o .claude/agents/architect-reviewer-complex.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/architect-reviewer-complex.md $ curl -fsSL --create-dirs -o .claude/agents/architect-reviewer.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/architect-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/implementer-complex.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/implementer-complex.md $ curl -fsSL --create-dirs -o .claude/agents/implementer-simple.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/implementer-simple.md $ curl -fsSL --create-dirs -o .claude/agents/implementer.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/implementer.md $ curl -fsSL --create-dirs -o .claude/agents/researcher-complex.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/researcher-complex.md $ curl -fsSL --create-dirs -o .claude/agents/researcher-simple.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/researcher-simple.md $ curl -fsSL --create-dirs -o .claude/agents/researcher.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/researcher.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer-complex.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/reviewer-complex.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/ui-reviewer-complex.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/ui-reviewer-complex.md $ curl -fsSL --create-dirs -o .claude/agents/ui-reviewer.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/ui-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/verifier.md https://raw.githubusercontent.com/tammai/ai-sdlc/main/agents/verifier.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} Skills (16)
Subagents (13)
| architect-reviewer-complex model: opus | [COMPLEX tier (risk tier L) — opus/medium] Adversarial reviewer for a system design (design.md) or an ADR, with fresh context — attacks capacity math, single points of failure, data ownership, consist |
| architect-reviewer model: sonnet | [NORMAL tier (risk tier M) — sonnet/high] Adversarial reviewer for a system design (design.md) or an ADR, with fresh context — attacks capacity math, single points of failure, data ownership, consiste |
| implementer-complex model: opus | [COMPLEX tier (risk tier L) — opus/medium] Implements an APPROVED plan.md for the active ai-sdlc change — exactly the files and order in the plan, tests as specified, `sdlc verify` green before report |
| implementer-simple model: sonnet | [SIMPLE tier (risk tier S) — sonnet/low] Implements an APPROVED plan.md for the active ai-sdlc change — exactly the files and order in the plan, tests as specified, `sdlc verify` green before reportin |
| implementer model: sonnet | [NORMAL tier (risk tier M) — sonnet/high] Implements an APPROVED plan.md for the active ai-sdlc change — exactly the files and order in the plan, tests as specified, `sdlc verify` green before reporti |
| researcher-complex model: opus | [COMPLEX tier (risk tier L) — opus/medium] Read-only codebase researcher for the spec and plan stages — maps where a change lands (files, call sites, data model, contracts, tests, conventions) and its |
| researcher-simple model: sonnet | [SIMPLE tier (risk tier S) — sonnet/low] Read-only codebase researcher for the spec and plan stages — maps where a change lands (files, call sites, data model, contracts, tests, conventions) and its b |
| researcher model: sonnet | [NORMAL tier (risk tier M) — sonnet/high] Read-only codebase researcher for the spec and plan stages — maps where a change lands (files, call sites, data model, contracts, tests, conventions) and its |
| reviewer-complex model: opus | [COMPLEX tier (risk tier L) — opus/medium] Independent code reviewer with fresh context — runs the bugs, security and compliance passes from REVIEW.md against the diff and the change's spec.md/plan.md |
| reviewer model: sonnet | [NORMAL tier (risk tier M) — sonnet/high] Independent code reviewer with fresh context — runs the bugs, security and compliance passes from REVIEW.md against the diff and the change's spec.md/plan.md, |
| ui-reviewer-complex model: opus | [COMPLEX tier (risk tier L) — opus/medium] Independent UI/UX reviewer with fresh context — audits changed screens against DESIGN.md, ui.md and the uiux craft rules through five lenses (accessibility, |
| ui-reviewer model: sonnet | [NORMAL tier (risk tier M) — sonnet/high] Independent UI/UX reviewer with fresh context — audits changed screens against DESIGN.md, ui.md and the uiux craft rules through five lenses (accessibility, p |
| verifier model: sonnet | Runs the app and checks the change actually works against plan.md before the session reports done. Read-only — reports, never fixes. Use after implementation passes `sdlc verify`, especially for UI, A |
Similar rigs
nicodiansk/turboflow
Opinionated engineering pipeline for Claude Code with 10 skills, 5 agents, 2 hooks. Replaces ad-hoc agent sprawl with brainstorm → story → plan → build → review → ship.
Orchestrator 1.9k tok ·
yoshpy-dev/ralph
Claude Code + Codex harness engineering — scaffold, upgrade, and run opinionated agent harnesses with deterministic hooks, evidence-backed review pipelines, and an autonomous multi-seat org runtime.
YOLO Cowboy 4.7k tok ·
itw-creative-works/workkit
The issue-pipeline workflow system as a Claude Code plugin: GitHub Issues as the work-item SSOT, labels as the pipeline, guard hooks, a manager crew, and the skills from build to ship.
Pragmatist 3.9k tok ·
jjanczur/tyran
A task conductor for Claude Code that gets better the more you use it. Multi-agent orchestration with an evidence contract that BLOCKS instead of asking, gates written as hooks rather than prompts, and a retrospective that learns your repo'
Orchestrator 3.2k tok ·