~ / rigs / BeastabTech / agent-workflow-kit

BeastabTech/agent-workflow-kit

Team workflow for coding agents: manager + workers, spawn limits, a closer that checks the finished feature, safety hooks, short project memory. Claude Code plugin + plain templates for Codex, Cursor, Gemini, Copilot. macOS, Linux, Windows.

↗ GitHub ★ 2 MIT updated 22d ago project Claude CodeCodexCursorGemini CLICopilot Claude plugin
share on X
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
CONTEXT TAX · EVERY TURN
~5.2k tokens
Moderate · median rig: 2.3k · breakdown
GUARDRAILS
1/5
No YOLO mode · details
Quality 18/100 Guardrails v2

Copy this rig

# review before running: this installs third-party code
$ npx degit BeastabTech/agent-workflow-kit/templates/adapters/claude/.claude ./rig-agent-workflow-kit  # 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 BeastabTech/agent-workflow-kit/skills/agent-comms .claude/skills/agent-comms
$ npx degit BeastabTech/agent-workflow-kit/skills/ai-feature .claude/skills/ai-feature
$ npx degit BeastabTech/agent-workflow-kit/skills/api-design .claude/skills/api-design
$ npx degit BeastabTech/agent-workflow-kit/skills/aws-ops .claude/skills/aws-ops
$ npx degit BeastabTech/agent-workflow-kit/skills/bug-intake .claude/skills/bug-intake
$ npx degit BeastabTech/agent-workflow-kit/skills/caveman .claude/skills/caveman
$ npx degit BeastabTech/agent-workflow-kit/skills/closeout .claude/skills/closeout
$ npx degit BeastabTech/agent-workflow-kit/skills/context-budget .claude/skills/context-budget
$ npx degit BeastabTech/agent-workflow-kit/skills/db-migration .claude/skills/db-migration
$ npx degit BeastabTech/agent-workflow-kit/skills/debug .claude/skills/debug
$ npx degit BeastabTech/agent-workflow-kit/skills/delegate .claude/skills/delegate
$ npx degit BeastabTech/agent-workflow-kit/skills/deploy .claude/skills/deploy
$ npx degit BeastabTech/agent-workflow-kit/skills/design-review .claude/skills/design-review
$ npx degit BeastabTech/agent-workflow-kit/skills/discover .claude/skills/discover
$ npx degit BeastabTech/agent-workflow-kit/skills/docs-sync .claude/skills/docs-sync
$ npx degit BeastabTech/agent-workflow-kit/skills/file-uploads .claude/skills/file-uploads
$ npx degit BeastabTech/agent-workflow-kit/skills/frontend-design .claude/skills/frontend-design
$ npx degit BeastabTech/agent-workflow-kit/skills/gcp-ops .claude/skills/gcp-ops
$ npx degit BeastabTech/agent-workflow-kit/skills/git-workflow .claude/skills/git-workflow
$ npx degit BeastabTech/agent-workflow-kit/skills/handoff .claude/skills/handoff
$ npx degit BeastabTech/agent-workflow-kit/skills/html-page .claude/skills/html-page
$ npx degit BeastabTech/agent-workflow-kit/skills/init .claude/skills/init
$ npx degit BeastabTech/agent-workflow-kit/skills/lessons .claude/skills/lessons
$ npx degit BeastabTech/agent-workflow-kit/skills/loop .claude/skills/loop
$ npx degit BeastabTech/agent-workflow-kit/skills/manager .claude/skills/manager
$ npx degit BeastabTech/agent-workflow-kit/skills/mobile-app .claude/skills/mobile-app
$ npx degit BeastabTech/agent-workflow-kit/skills/peer-review .claude/skills/peer-review
$ npx degit BeastabTech/agent-workflow-kit/skills/plan .claude/skills/plan
$ npx degit BeastabTech/agent-workflow-kit/skills/responsive-ui .claude/skills/responsive-ui
$ npx degit BeastabTech/agent-workflow-kit/skills/safe-change .claude/skills/safe-change
$ npx degit BeastabTech/agent-workflow-kit/skills/security-audit .claude/skills/security-audit
$ npx degit BeastabTech/agent-workflow-kit/skills/self-review .claude/skills/self-review
$ npx degit BeastabTech/agent-workflow-kit/skills/skill-testing .claude/skills/skill-testing
$ npx degit BeastabTech/agent-workflow-kit/skills/state-keeper .claude/skills/state-keeper
$ npx degit BeastabTech/agent-workflow-kit/skills/tdd .claude/skills/tdd
$ npx degit BeastabTech/agent-workflow-kit/skills/team-message .claude/skills/team-message
$ npx degit BeastabTech/agent-workflow-kit/skills/team-status .claude/skills/team-status
$ npx degit BeastabTech/agent-workflow-kit/skills/timezones-i18n .claude/skills/timezones-i18n
$ npx degit BeastabTech/agent-workflow-kit/skills/webapp-testing .claude/skills/webapp-testing
$ curl -fsSL --create-dirs -o .claude/agents/build-fixer.md https://raw.githubusercontent.com/BeastabTech/agent-workflow-kit/production/agents/build-fixer.md
$ curl -fsSL --create-dirs -o .claude/agents/closer.md https://raw.githubusercontent.com/BeastabTech/agent-workflow-kit/production/agents/closer.md
$ curl -fsSL --create-dirs -o .claude/agents/doc-updater.md https://raw.githubusercontent.com/BeastabTech/agent-workflow-kit/production/agents/doc-updater.md
$ curl -fsSL --create-dirs -o .claude/agents/explorer.md https://raw.githubusercontent.com/BeastabTech/agent-workflow-kit/production/agents/explorer.md
$ curl -fsSL --create-dirs -o .claude/agents/planner.md https://raw.githubusercontent.com/BeastabTech/agent-workflow-kit/production/agents/planner.md
$ curl -fsSL --create-dirs -o .claude/agents/qa-tester.md https://raw.githubusercontent.com/BeastabTech/agent-workflow-kit/production/agents/qa-tester.md
$ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/BeastabTech/agent-workflow-kit/production/agents/reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/security-reviewer.md https://raw.githubusercontent.com/BeastabTech/agent-workflow-kit/production/agents/security-reviewer.md
$ curl -fsSL --create-dirs -o .claude/agents/test-writer.md https://raw.githubusercontent.com/BeastabTech/agent-workflow-kit/production/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 (39)

Subagents (9)

build-fixer
model: inherit
Fixes failing type-check, build, lint or test-setup errors with the smallest change that makes the checks pass, then re-runs them. Use when a build, type-check or lint run fails and the cause is mecha
closer
model: inherit
Last agent of a piece of work. Hunts for what is missing or broken in the finished feature, fixes clear low-risk items, asks the human only for real decisions, re-tests, and sends the manager a Closeo
doc-updater
model: haiku
Brings project docs in line with a finished change - the feature file in docs/context/features, module map and owners, conventions, env example, API docs, STATE hand-off notes - and marks replaced tex
explorer
model: haiku
Read-only codebase mapper. Use to locate code, trace a feature across layers and docs, list what depends on a function, route, field or event, or answer "where is X" questions without filling the main
planner
model: inherit
Read-only planner. Turns a goal into numbered tasks with owned files, order, dependencies, the cheapest model per task, and an agent-count estimate to check against spawn limits. Use at the start of m
qa-tester
model: inherit
QA agent that tests a feature like a user would - runs the app or API, walks the happy path, edge cases, permissions and error states, and files clear bug reports into .agents/BUGS.md. Use after a fea
reviewer
model: inherit
Strict read-only code reviewer. Use after a change and before saying done, or when asked to review a diff, branch or another agent's work. Checks correctness, security, concurrency, edge cases, tests,
security-reviewer
model: inherit
Read-only security reviewer. Use before merging changes that touch auth, permissions, user input, file uploads, payments, secrets, external calls, infrastructure or dependencies, or when asked for a s
test-writer
model: inherit
Writes tests from acceptance criteria or a bug report - a failing test first, then edge cases, permissions and old-flow smoke tests - and runs them. Use before an implementation (TDD), for a bug's reg

Cursor rules (1)

agent-workflow · always

Similar rigs

copied ✓