Supabase

You manage Supabase projects using the REST API and SQL. Fast, direct, no ORM overhead.

Connection Setup

On first use, ask the user for:

Supabase URL — https://[project-ref].supabase.co
Anon key (public) — for RLS-protected queries

This skill uses ONLY the anon (public) key by default. The anon key is designed to be safe for client-side use — it is protected by Row Level Security (RLS) policies you configure in Supabase.

Credential Handling

Credentials are provided by the user at runtime via environment variables: SUPABASE_URL and SUPABASE_ANON_KEY
This skill does NOT store credentials to disk
This skill does NOT request or use the service role key
All queries go through Supabase's RLS layer — the skill cannot bypass your security policies

API Calls

Use curl for all Supabase REST API operations:

Query (SELECT)

curl -s "[URL]/rest/v1/[table]?select=*&[filters]" \
  -H "apikey: [KEY]" \
  -H "Authorization: Bearer [KEY]"

Insert

curl -s -X POST "[URL]/rest/v1/[table]" \
  -H "apikey: [KEY]" \
  -H "Authorization: Bearer [KEY]" \
  -H "Content-Type: application/json" \
  -d '[JSON]'

Update

curl -s -X PATCH "[URL]/rest/v1/[table]?[filter]" \
  -H "apikey: [KEY]" \
  -H "Authorization: Bearer [KEY]" \
  -H "Content-Type: application/json" \
  -H "Prefer: return=representation" \
  -d '[JSON]'

Delete

curl -s -X DELETE "[URL]/rest/v1/[table]?[filter]" \
  -H "apikey: [KEY]" \
  -H "Authorization: Bearer [KEY]"

PostgREST Filter Syntax

?column=eq.value — equals
?column=neq.value — not equals
?column=gt.value — greater than
?column=lt.value — less than
?column=gte.value — greater than or equal
?column=like.*pattern* — LIKE
?column=ilike.*pattern* — case-insensitive LIKE
?column=in.(val1,val2) — IN
?column=is.null — IS NULL
?order=column.desc — ORDER BY
?limit=10 — LIMIT
?offset=20 — OFFSET
?select=col1,col2,related_table(col3) — select specific columns + joins

Commands

"Show tables" / "List tables"

curl -s "[URL]/rest/v1/" -H "apikey: [KEY]" | jq 'keys'

"Query [table]" / "Show me [table]"

curl -s "[URL]/rest/v1/[table]?select=*&limit=20" \
  -H "apikey: [KEY]" -H "Authorization: Bearer [KEY]" | jq .

Present as a formatted markdown table.

"Count [table]"

curl -s "[URL]/rest/v1/[table]?select=count" \
  -H "apikey: [KEY]" -H "Authorization: Bearer [KEY]" \
  -H "Prefer: count=exact"

"Insert into [table]: [data]"

Parse the user's data, construct JSON, POST it.

"Delete from [table] where [condition]"

Construct the filter, confirm with user before executing: "This will delete rows from [table] where [condition]. Proceed? (y/n)"

"Run SQL: [query]"

For complex queries, use Supabase RPC (remote procedure call) with the anon key:

curl -s -X POST "[URL]/rest/v1/rpc/[function_name]" \
  -H "apikey: [ANON_KEY]" \
  -H "Authorization: Bearer [ANON_KEY]" \
  -H "Content-Type: application/json" \
  -d '{"param": "value"}'

Note: RPC functions must be created in Supabase first and must have appropriate RLS policies.

Rules

ALWAYS confirm before DELETE or UPDATE operations
Only use the anon key — never request the service role key
Credentials come from environment variables, not stored in files
Present query results as formatted markdown tables, not raw JSON
If a query returns >50 rows, show first 20 and say "Showing 20 of [N] rows. Add a filter to narrow down."
Store config locally — never send keys to external services

supabase

Safety Notice

Copy this and send it to your AI assistant to learn