adyusuf/claude-code-standards
The live Claude Code rule set I ship production software with — 33 rules, 22 standards, 14 agent roles, each traced to the measurement or incident that produced it.
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 adyusuf/claude-code-standards/agents ./rig-claude-code-standards/agents $ npx degit adyusuf/claude-code-standards/commands ./rig-claude-code-standards/commands $ npx degit adyusuf/claude-code-standards/skills ./rig-claude-code-standards/skills $ npx degit adyusuf/claude-code-standards/hooks ./rig-claude-code-standards/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 adyusuf/claude-code-standards/skills/board-plan .claude/skills/board-plan $ npx degit adyusuf/claude-code-standards/skills/config .claude/skills/config $ npx degit adyusuf/claude-code-standards/skills/software-standards .claude/skills/software-standards $ npx degit adyusuf/claude-code-standards/skills/test-inventory .claude/skills/test-inventory $ npx degit adyusuf/claude-code-standards/skills/working-mode .claude/skills/working-mode
$ curl -fsSL --create-dirs -o .claude/agents/analyst.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/analyst.md $ curl -fsSL --create-dirs -o .claude/agents/architect.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/architect.md $ curl -fsSL --create-dirs -o .claude/agents/coverage-auditor.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/coverage-auditor.md $ curl -fsSL --create-dirs -o .claude/agents/data.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/data.md $ curl -fsSL --create-dirs -o .claude/agents/designer.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/designer.md $ curl -fsSL --create-dirs -o .claude/agents/developer.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/developer.md $ curl -fsSL --create-dirs -o .claude/agents/devops.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/devops.md $ curl -fsSL --create-dirs -o .claude/agents/doc-writer.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/doc-writer.md $ curl -fsSL --create-dirs -o .claude/agents/e2e-writer.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/e2e-writer.md $ curl -fsSL --create-dirs -o .claude/agents/observability.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/observability.md $ curl -fsSL --create-dirs -o .claude/agents/product-manager.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/product-manager.md $ curl -fsSL --create-dirs -o .claude/agents/qa.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/qa.md $ curl -fsSL --create-dirs -o .claude/agents/security.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/security.md $ curl -fsSL --create-dirs -o .claude/agents/test-writer.md https://raw.githubusercontent.com/adyusuf/claude-code-standards/prod/agents/test-writer.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 (5)
Subagents (14)
| analyst model: sonnet | Investigates how existing code works. Use it for "where is this defined", "how does this flow work", "how many places use this". Reads a lot, returns SHORT. Does NOT change code. |
| architect model: opus | Produces an implementation plan — which file, what changes, in what order. Use it before any code is written on work that touches 10+ files. Does not write code. |
| coverage-auditor model: sonnet | Measures line coverage for every codebase and audits the 80% threshold and the honesty of the denominator (#29). Does NOT write tests and does not change product code. |
| data model: opus | Reviews and plans schema, migrations, indexes, transactions and data migration. Audits backward compatibility and the risk of data loss. Does NOT run migrations. |
| designer model: sonnet | Produces interface/UX decisions — flow, states, accessibility, empty/error states. Use it when a screen or component is being designed. |
| developer model: opus | Writes an isolated piece of code with a CLEAR contract (one file, defined signature, specified behaviour). Do NOT use it for cross-layer work or work that needs exploration. |
| devops model: sonnet | Reviews and prepares CI, deploy, gate, backup and environment configuration work. Does NOT run deploys. |
| doc-writer model: haiku | Updates CLAUDE.md and docs/. Writes permanent decisions into the right file. Does NOT change code. |
| e2e-writer model: sonnet | Writes Playwright/Maestro e2e specs and classifies failing runs. Runs ONLY against the test environment; never runs dev code and never changes product code. |
| observability model: sonnet | Reviews the logging, metrics, tracing, alerting and performance axes. Finds what will be invisible in production. Does NOT change code. |
| product-manager model: sonnet | Turns a request into scope, acceptance criteria and edge cases. Use it BEFORE any code is written on vague or broad requests. Does not write code. |
| qa model: opus | Reviews a change for correctness, security, backward compatibility and test coverage. The first review pass in modes C/D/E. Does NOT change code. |
| security model: opus | Reviews the change and the repository for security — OWASP, secret leakage, authorization, dependency CVEs, gate integrity. Does NOT change code and does NOT run scans. |
| test-writer model: sonnet | Writes tests for a specified behaviour. Use it for isolated test work with a CLEAR scope. Does not change product code. |
Slash commands (4)
/apply-project-standards/panels/prod-readiness/standards-check
Similar rigs
ThinhHoang1/claude-code-agent-team
An AI dev team for Claude Code agent teams: 8 role agents, a HANDOFF protocol and 3 hook quality gates — with the TeammateIdle infinite-loop bug fixed and regression-tested.
Orchestrator 1.6k tok · · Q15
FCabiddu/pocket-it
Agent system for Claude Code: analyst, architect, task-writer, developer, reviewer, security and tester, with hooks and scripts that enforce the gates
Orchestrator 3.2k tok · · Q5
secretj/secretj-claude-config
개인 Claude Code dotfiles — agents, commands, skills, hooks, memory, MCP 템플릿. clone + install.sh 로 환경 복원.
Orchestrator 1.8k tok · · Q15
danielfarnose/claude-code-dev-loop
A multi-agent development loop for Claude Code: five specialized subagents plan, implement, verify and merge — with a QA gate that cannot edit the code it judges.
Orchestrator 1.2k tok · · Q0