BinarySword/kiro-project-template
Opinionated Kiro project template with 40+ steering docs, 34 agent hooks, sample specs, and a CLI scaffolder — get a production-ready .kiro/ setup in seconds.
ARCHETYPE
MCP Hoarder
Six or more MCP servers wired in. Every tool, always in context.
Copy this rig
# review before running: this installs third-party code $ claude mcp add supabase -e SUPABASE_ACCESS_<redacted> -- npx -y @supabase/mcp-server-supabase@latest --read-only $ claude mcp add fetch -- uvx mcp-server-fetch $ claude mcp add git -- uvx mcp-server-git --repository . $ claude mcp add chrome -- npx -y chrome-devtools-mcp@latest $ claude mcp add docker -- uvx mcp-server-docker $ claude mcp add time -- uvx mcp-server-time $ claude mcp add memory -- npx -y @modelcontextprotocol/server-memory $ claude mcp add sequential-thinking -- npx -y @modelcontextprotocol/server-sequential-thinking
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.
$ claude mcp add supabase -e SUPABASE_ACCESS_<redacted> -- npx -y @supabase/mcp-server-supabase@latest --read-only $ claude mcp add fetch -- uvx mcp-server-fetch $ claude mcp add git -- uvx mcp-server-git --repository . $ claude mcp add chrome -- npx -y chrome-devtools-mcp@latest $ claude mcp add docker -- uvx mcp-server-docker $ claude mcp add time -- uvx mcp-server-time $ claude mcp add memory -- npx -y @modelcontextprotocol/server-memory $ claude mcp add sequential-thinking -- npx -y @modelcontextprotocol/server-sequential-thinking
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 *)"
]
}
} MCP servers (8)
| server | source | est. tokens |
|---|---|---|
| Supabase env: SUPABASE_ACCESS_TOKEN | npm | 7.0k |
| Fetch | pypi | 450 |
| Git | pypi | 3.2k |
| Chrome DevTools MCP · "chrome" | npm | 9.0k |
| mcp-server-docker · "docker" | pypi | 2.5k |
| Time | pypi | 500 |
| Memory (knowledge graph) · "memory" | npm | 2.6k |
| Sequential Thinking · "sequential-thinking" | npm | 900 |
Hooks (40)
| event | matcher | runs |
|---|---|---|
| Stop | * | Task complete. If source code was modified: 1. Run `trunk check --fix <modified-files>` for lint/format issues 2. Run type checking (e.g. `pnpm tsc --noEmit`) 3. Run affected tests (e.g. `pnpm test -- --run`) If only markdown or config fi |
| PostFileSave | **/*.test.ts,**/*.test.tsx,**/*.spec.ts,**/*.property.test.ts,e2e/**/*.spec.ts | Test file modified. Verify: 1. Follows AAA pattern (Arrange-Act-Assert) 2. No anti-patterns (hard-coded waits, test.only, empty catch) 3. Edge cases covered (null, empty, error states) 4. Property tests have numRuns >= 100 and timeout >= 3 |
| PostFileSave | .env*,docker-compose*.yml,*.config.ts,*.config.mjs,next.config.* | Config file modified. Check for: 1. Exposed secrets or API keys (must use env vars) 2. Insecure default values 3. Missing documentation in .env.example 4. Security best practices for the config type |
| PostFileSave | **/server/api/routers/**/*.ts,**/trpc/**/*.ts,**/api/routers/**/*.ts | Check the saved tRPC router file for common issues: 1. Every procedure must have `.input()` with a Zod schema — no unvalidated inputs 2. Mutations must use `protectedProcedure` unless explicitly public 3. Procedures should not contain busi |
| PostFileSave | **/*.functions.ts,**/routes/**/*.tsx | Check the saved file for TanStack Start server function issues: 1. Server functions created with `createServerFn` must have `.inputValidator()` with a Zod schema for any function that accepts input 2. Server-only logic (DB queries, secrets |
| PostFileSave | src/app/api/**/*.ts,src/api/**/*.ts,packages/*/src/api/**/*.ts,openapi.yaml,swagger.yaml | API code modified. Verify: 1. Run affected API tests 2. Check if API documentation needs updating 3. Verify error response format matches standard 4. Check for breaking changes to existing consumers |
| PostFileSave | .kiro/specs/**/*.md | Spec file modified. Validate: 1. All file references (#[[file:...]]) resolve correctly 2. Required sections are present per template 3. Task numbering is sequential and consistent 4. Requirement cross-references (R1, R2, etc.) are valid 5. |
| PostFileSave | src/lib/db/schema/*.ts,src/lib/db/schema.ts,packages/*/db/schema/*.ts | Database schema modified. Consider: 1. Generate a migration file for the schema change 2. Verify the migration is reversible 3. Update TypeScript types if needed 4. Check for data migration requirements Remember: Always generate migration |
| PostFileSave | package.json,packages/*/package.json | Dependencies modified. Consider: 1. Run audit for known vulnerabilities 2. Check for outdated critical packages 3. Regenerate lockfile if needed 4. Review new dependencies for security and bundle size impact |
| PostFileSave | **/*.openapi.yml,**/*.openapi.yaml,**/*.openapi.json,**/openapi.*,**/*.graphql,**/*.gql,**/schema.graphql | API schema file modified. Validate: 1. Schema syntax is valid (OpenAPI 3.1 or GraphQL SDL) 2. All referenced types and schemas are defined 3. Required fields are marked correctly 4. Response codes follow REST conventions (if OpenAPI) 5. Na |
| PostFileSave | **/*.ts,**/*.tsx,**/*.js,**/*.jsx,**/*.json,**/*.yml,**/*.yaml,**/*.env* | Quick scan the modified file for hardcoded secrets. Check for: 1. API keys (patterns like `sk_live_`, `AKIA`, `ghp_`, `xox`) 2. Passwords or tokens assigned to variables 3. Connection strings with embedded credentials 4. Base64-encoded sec |
| Manual | * | Run comprehensive project health check. ## Checks to Perform ### 1. Code Quality - Run `trunk check` on entire codebase - Run type checking (`pnpm tsc --noEmit`) - Check for TODO/FIXME comments ### 2. Tests - Run `pnpm test -- --run` for |
| Manual | * | Check database health and performance. ## Checks to Perform 1. Verify database connection is active 2. List tables and check for structural issues 3. Check applied migrations are up to date 4. Review for pending schema changes ## Output | |
| Manual | * | Generate a conventional commit message from the current git diff. 1. Analyse staged changes (or working tree if nothing staged) 2. Determine commit type: feat, fix, docs, style, refactor, test, chore, perf, build, ci 3. Identify scope from |
| Manual | * | Run a performance analysis on the codebase. ## Check For 1. **Bundle size** — Large imports, missing tree-shaking, unnecessary dependencies 2. **React rendering** — Missing memoisation where needed, unnecessary re-renders, missing Suspens |
| Manual | * | Review all markdown documentation files (README.md, docs/**/*.md) for: 1. Spelling errors 2. Grammar issues 3. Inconsistent terminology 4. Broken links or references 5. Australian English compliance (optimise, behaviour, colour, analyse, l |
| Manual | * | Audit all React components (src/**/*.tsx) for accessibility issues: 1. Missing alt text on images 2. Missing aria-label on interactive elements without visible text 3. Non-semantic HTML (div with onClick instead of button) 4. Missing keybo |
| Manual | * | Scan the codebase for dead code: 1. Unused exports — functions, types, or constants exported but never imported elsewhere 2. Unused imports — imported symbols not referenced in the file 3. Unused variables — declared but never read 4. Unre |
| PostTaskExecution | * | trunk check --fix --all |
| Manual | * | Scan all source files in `src/` for TODO, FIXME, HACK, and XXX comments. For each one found: 1. File path and line number 2. The comment text 3. Age estimate (check git blame if possible) 4. Severity: FIXME/HACK = high, TODO = medium 5. Su |
| Manual | * | Analyse the project's bundle size and dependency weight: 1. Check `package.json` for heavy dependencies (known large packages) 2. Look for imports that pull in entire libraries instead of specific modules 3. Check for duplicate dependencie |
| Manual | * | Generate a changelog entry from recent git commits. 1. Read commits since the last tag (or last 50 commits if no tags) 2. Group by conventional commit type: - **Added** (feat) - **Fixed** (fix) - **Changed** (refactor, perf) - |
| Manual | * | Scan `package.json` dependencies for license compatibility: 1. Read all dependencies and devDependencies 2. Check for known problematic licences: - GPL/AGPL in an MIT/Apache project (copyleft contamination) - SSPL (Server Side Public |
| Manual | * | Identify stale git branches that can be cleaned up: 1. Run `git branch --merged main` to find branches already merged into main 2. Exclude `main`, `develop`, and any release branches 3. For each candidate, show: - Branch name - Last |
| Manual | * | Use steering doc `32-feature-planning.md` if available. Review the current feature idea, draft spec, or implementation plan before code is written. Required outputs: 1. Premise challenge 2. Scope mode recommendation (Expansion, Hold Scope, |
| Manual | * | Use steering doc `33-qa-methodology.md` if available. Perform a structured QA pass in the best available mode: Diff-Aware, Full, or Quick. Rules: - If browser tooling is available, use it and gather evidence - If browser tooling is unavail |
| Manual | * | Use steering doc `34-investigation.md` if available. Investigate the current bug or failure using a reproduce -> isolate -> hypothesise -> confirm -> implement -> verify flow. Rules: - Do not edit production code until you have a root-caus |
| Manual | * | Use steering doc `73-document-release.md` if available. Compare the current diff or recent shipped work against project documentation and update drifted docs. Check at least: - README.md - docs/README.md - docs/ARCHITECTURE.md - docs/API.m |
| Manual | * | Use steering doc `74-security-audit.md` if available. Run a structured security review in Daily or Comprehensive mode depending on scope and risk. Review at least: 1. Secrets and sensitive data exposure 2. Authentication and authorisation |
| Manual | * | Use steering doc `35-safety-guardrails.md` if available. Activate safety mode for the current conversation. Required behaviour: 1. State the active edit boundary at the start of the task; if unclear, ask the user to name the directory or m |
Kiro (40 steering · 2 specs · 48 hook files)
00-core-rules · always · 557 tok01-product · always · 364 tok02-tech · always · 1.2k tok03-structure · always · 711 tok10-dev-code-style · always · 438 tok11-dev-error-handling · filematch12-dev-testing · filematch20-workflow-git · always · 865 tok21-workflow-task-completion · always · 554 tok30-hooks-guide · manual31-kiro-best-practices · manual32-feature-planning · manual33-qa-methodology · manual34-investigation · manual35-safety-guardrails · manual40-security · always · 628 tok41-docker · filematch42-react-components · filematch43-mcp · always · 605 tok44-authorisation · filematch45-api-standards · filematch46-logging · filematch47-database · filematch48-documentation · filematch49-cicd · filematch50-code-review · manual50a-spec-creation · filematch51-accessibility · filematch52-bug-tracking · always · 1.0k tok53-nextjs · filematch54-state-management · filematch55-tailwind-shadcn · filematch56-i18n · filematch57-performance · filematch58-env-variables · filematch59-realtime · filematch60-t3-stack · filematch61-t4-stack · filematch62-tanstack · filematch63-stack-selection · manual
Specs: _BUGFIX_TEMPLATE, _TEMPLATE (loaded on demand, not counted in context tax)
MCP config: 44 auto-approved tool(s) · disabled (not counted): supabase, docker
How Kiro files are scored: methodology.
Similar rigs
Cordycepsers/vscode-claude-setup-2025
Complete configuration for Python, Node.js, React, and TypeScript development with VS Code, Claude Desktop, and Claude Code
MCP Hoarder 53.0k tok · · Q15
jacks0n/dotfiles
My dotfiles
MCP Hoarder 34.2k tok · · Q0
cbolden15/claude-code-config-manager
A powerful configuration management tool for Claude Code and Auto-Claude projects. CCM serves as the single source of truth for agent personas, MCP servers, model profiles, and project settings across multiple machines.
Pragmatist 15.8k tok · · Q0
samjtro/claude
personal claude code config
Pragmatist 7.2k tok · · Q0