stack-defaults
Use when choosing or confirming tech stack for a product or feature: frontend framework, database, auth, email, payments, hosting, search, or agent runtime.
Skill by AVIDS2
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/avids2-stack-defaults
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: frontend, backend, testing, security
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/stack-defaults
What's inside
- Skill name
stack-defaults- Description
- Use when choosing or confirming tech stack for a product or feature: frontend framework, database, auth, email, payments, hosting, search, or agent runtime. Applies AVID defaults (React/Vue, SQLite→Postgres/Supabase, Resend, Stripe, Playwright, Tavily, Context7) and records exceptions. Use during product-bootstrap or any greenfield/stack debate.
- License
- MIT
- Instructions
- 63 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- files: referenced but missing: references/stack-table.md
Files (1)
- SKILL.md
SKILL.md (first lines)
# Stack Defaults Opinionated **defaults** for AVID projects. Easy to override — record a short reason when you do. Scene beats table. ## Default Stack Table | Layer | Default | Notes | |-------|---------|-------| | Frontend | **React or Vue** | Match existing repo; greenfield prefer React+TS unless user is Vue-native | | UI quality | Design skills / impeccable patterns | No generic AI slop UI | | Web QA | **Playwright** | Prefer a11y snapshots over pure screenshots | | Email | **Resend** | Transactional + registration flows | | Payments | **Stripe** | Don't hand-roll billing | | Data (MVP) | **SQLite** | Local/dev/single-node first | | Data (prod) | **Postgres** via **Supabase** when it fits | Auth/storage/realtime optional | | Search (web) | **Tavily** | Default research path | | Search (vertical) | **AnySearch** | Domain/structured retrieval | | Lib docs | **Context7** | New/fast-moving APIs — no memory guessing | | Workflow agents | **LangGraph** + docs/skills | Only when workflow/HITL graph is real — see `agent-framework-choice` | | Local harness agents | **Pi SDK / pi-ai / embed Pi** | Local/desktop/workspace agents only — **not** default for ordinary web FE/BE | | **Model API protocol** (agent products) | **Chat Completions** `/v1/chat/completions` | Widest multi-model/gateway/Pi `openai-completions` support; Responses/Messages as optional extras | | Minimalism | **ponytail** | Lazier working solution | | Process (Claude Code) | **Matt Pocock skills** | Replaces Superpowers on CC; not for Codex/OpenCode | | Frontend craft | **Impeccable** + **taste-skill** (`design-taste-frontend`) | Pair: taste direction + Impeccable polish/audit | | Hosting (personal) | Existing VPS + Docker + reverse proxy pattern | Follow user infra notes | ## Decision Rules 1. **Repo / scene wins**: existing stack and deployment target override the table 2. **MVP data**: SQLite until multi-writer, managed auth need, or clear prod path 3. **Managed beats yak-shave** when it fits: Supabase/Stripe/Resend over custom SMTP + homegrown billing 4. **Docs before code** on fast-moving APIs (Context7 / official) 5. **One frontend framework** per product surface unless there's a real reason 6. **Agent runtime is scene-based** — never "always Pi" or "always LangGraph" 7. **Agent model wire protocol** — prefer Chat Completions for multi-provider; don't lock P0 to Responses-only unless product is OpenAI-only ## Exception Log Format When overriding a default, record: ```markdown - Layer: <e.g. email> - Default: Resend - Chosen: <X> - Reason: <constraint> ``` Put exceptions in `docs/spec.md` or market/bootstrap package. ## Anti-Patterns - Postgres-on-day-1 for a solo MVP with no multi-user load - Custom payment state machines instead of Stripe - Guessing LangGraph/Pi APIs without docs - Installing both React and Vue "for flexibility" - Building email sending with raw SMTP "to save money" without cost math ## Related - `product-bootstrap`, `agent-framework-choice`, `install-dont-reinvent` - `references/stack-table.md`
From AVIDS2/avid-skill, MIT.
More from AVIDS2/avid-skill
- 0.502score
using-avid
AVIDS2
Use when starting product, feature, build, stack, research, agent-framework, or implementation work.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
multi-agent-deliberation
AVIDS2
Use for high-stakes decisions: architecture forks, build-vs-buy, competitive response, major refactors, or when the user wants multi-agent debate.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
agent-framework-choice
AVIDS2
Use when choosing or implementing agent runtimes: LangGraph workflows, Pi / pi-ai / embedded Pi harness agents, thin provider SDK loops, model API wire…
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
product-bootstrap
AVIDS2
Prefer before greenfield products, new verticals, or "做个X/从零做" work.
SkillCFreshness: C★ 1Claude Code +9 - 0.493score
knowledge-architecture
AVIDS2
Use when designing knowledge bases, RAG, wiki, knowledge graphs, memory for agents, or "how should we store docs for humans vs agents".
SkillCFreshness: C★ 1Claude Code +9 - 0.489score
install-dont-reinvent
AVIDS2
Prefer when a required CLI, library, MCP, or skill is missing.
SkillCFreshness: C★ 1Claude Code +9 - 0.489score
research-before-build
AVIDS2
Prefer before self-implementing a non-trivial library, service, or subsystem.
SkillCFreshness: C★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Jul 24, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/AVIDS2/avid-skill
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from AVIDS2
- 0.761score

Memorix
AVIDS2
Local-first project memory with legacy MCP and 2026 discovery compatibility.
ConnectorAFreshness: ALocal★ 857Claude Code +5 - 0.502score
using-avid
AVIDS2
Use when starting product, feature, build, stack, research, agent-framework, or implementation work.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
multi-agent-deliberation
AVIDS2
Use for high-stakes decisions: architecture forks, build-vs-buy, competitive response, major refactors, or when the user wants multi-agent debate.
SkillCFreshness: C★ 1Claude Code +9 - 0.497score
agent-framework-choice
AVIDS2
Use when choosing or implementing agent runtimes: LangGraph workflows, Pi / pi-ai / embedded Pi harness agents, thin provider SDK loops, model API wire…
SkillCFreshness: C★ 1Claude Code +9 - 0.494score
product-bootstrap
AVIDS2
Prefer before greenfield products, new verticals, or "做个X/从零做" work.
SkillCFreshness: C★ 1Claude Code +9 - 0.493score
knowledge-architecture
AVIDS2
Use when designing knowledge bases, RAG, wiki, knowledge graphs, memory for agents, or "how should we store docs for humans vs agents".
SkillCFreshness: C★ 1Claude Code +9
Cite this page
stack-defaults skill. Agent Mart. https://agentmart.dev/skills/avids2-stack-defaults (last updated 2026-10-24)