Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/luanpdd/kit-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/luanpdd/kit-mcp/supabase)<a href="https://agentmods.dev/commands/luanpdd/kit-mcp/supabase"><img src="https://agentmods.dev/badge/commands/luanpdd/kit-mcp/supabase.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00047 | $0.05238 |
| Opus 5 | $0.00023 | $0.02619 |
| Sonnet 5 | $0.00009 | $0.01048 |
| Haiku 4.5 | $0.00005 | $0.00524 |
Grade A, and why
supabase scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 239 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Faz dispatch via Task(subagent_type=supabase-...) para o agent especializado correto. É o único ponto de chain de agents Supabase — agents permanecem função pura (anti-pitfall A10 de v1.8).
Princípio canônico v1.23: Agents não-Supabase pensam/planejam; agents Supabase materializam/hardenam; ninguém descarta upstream.
Cria/Atualiza: o que cada agent invocado cria/atualiza (migrations, schemas, functions, etc.) — com RLS auto-injetada no output via handoff cooperativo com supabase-rls-hardener em CREATE TABLE.
Após: o usuário tem o output do agent (plano, código, SQL hardenado, ou veredito GO/STRENGTHEN/REWRITE).
<execution_context>
Skills consultadas pelos agents: kit/skills/supabase-*/SKILL.md + kit/skills/_shared-supabase/glossary.md (Phase 25).
Agents disponíveis: kit/agents/supabase-*.md (Phase 26) + kit/agents/schema-checker.md (existente).
</execution_context>
Subcomandos suportados (sinônimos PT-BR/EN):
| Subcomando | Sinônimos | Agent dispatched |
|---|---|---|
arquiteto |
architect, arch |
supabase-architect |
migration |
migrar, migrate |
supabase-migration-writer (v1.23: auto-chain cooperativo com hardener em CREATE TABLE) |
rls |
— | supabase-rls-writer (v1.23: GRANTs + IS NOT NULL + views security_invoker) |
hardener |
harden, endurecer |
supabase-rls-hardener (v1.23 canonical materializer — recebe draft via Task) |
column |
coluna, col-priv |
supabase-column-privileges-writer (v1.24 canonical materializer column-level — recebe spec via Task) |
rbac |
roles, permissions, claims |
supabase-rbac-implementer (v1.25 canonical materializer Custom Claims & RBAC via Auth Hook — recebe spec via Task) |
role |
papel, roles-pg |
supabase-roles-implementer (v1.26 canonical materializer Postgres Roles — recebe spec via Task; system access) |
edge |
edge-function, function, funcao |
supabase-edge-fn-writer (v1.30: 2026 patterns — withSupabase, deno.json, config.toml) |
test |
testar, tests, deno-test |
supabase-edge-fn-tester (v1.30 — gera Deno tests para função existente) |
mcp |
mcp-server, mcp-lite |
supabase-edge-fn-writer com pattern=mcp-server |
ai |
ai-session, embeddings-builtin, gte-small, ollama |
supabase-edge-fn-writer com pattern=rag-embeddings (Supabase.ai.Session) |
wasm |
wasm-module |
supabase-edge-fn-writer com pattern=wasm + static_files config.toml |
websocket |
ws, realtime-ws |
supabase-edge-fn-writer com pattern=websocket + per_worker |
realtime |
tempo-real |
supabase-realtime-implementer |
auth |
autenticacao, auth-ssr |
supabase-auth-bootstrapper |
social |
oauth, login-social |
supabase-social-auth-implementer (v1.32 — social login Google/GitHub/Apple/Facebook/LinkedIn + custom OAuth/OIDC) |
mfa |
2fa, totp |
supabase-mfa-implementer (v1.32 — MFA TOTP/Phone + enforcement RLS por AAL) |
hooks |
auth-hook, hook |
supabase-auth-hook-writer (v1.32 — materializa os 6 Auth Hooks Postgres/HTTP) |
oauth-server |
mcp-auth, idp |
supabase-oauth-server-implementer (v1.32 — Supabase como OAuth 2.1/OIDC identity provider + MCP authentication) |
sso |
saml, enterprise-sso |
supabase-sso-saml-architect (v1.32 — Enterprise SSO SAML 2.0 + multi-tenant) |
storage |
armazenamento |
supabase-storage-implementer |
rag |
pgvector, embeddings |
supabase-edge-fn-writer com pattern=rag-embeddings |
cron |
queues, pgmq, background |
supabase-edge-fn-writer com pattern=cron-pgmq |
check |
validar, validate |
schema-checker (validação pré-migration) |
help |
ajuda, ? |
exibe esta tabela inline |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 239 lines · 47 tokens per session scan A 581c0e3fcd57
supabase is a command published in the GitHub repository luanpdd/kit-mcp (1 stars, last pushed yesterday), licensed MIT. It adds 47 tokens to every session and 5,238 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other commands, from other repositories
init
Initialize configurations for Supabase local development.
toh-connect
Connect app to Supabase backend with schema and RLS policies.
model-create
Create a new Django model following best-practice conventions.
services
Manage the connect service registry. Register API services and database connections, store credentials, set read-only vs write access, manage the destination allowlist, and report trust status.
p3-data-model
Designs the Data Model with entities and relationships as well as the API design with interface specification. The result is DATAMODEL.md and APISPEC.md as the binding foundation for implementation.
crud
Generates a complete CRUD domain in the house pattern (NestJS or Express + Prisma — picked by the project's real stack, multi-tenant).