* fix(sendgrid): fix active field coercion, add pagination, tighten output typing - Fix active field for create_template_version being sent as the string "true"/"false" instead of the SendGrid-required int 0/1 - Add missing authMode: ApiKey on SendGridBlock - Add pageToken/nextPageToken pagination support to list_templates and list_all_lists (SendGrid page_token cursor, parsed from _metadata.next) - Fix nullable output fields to use ?? null / ?? [] with optional: true across get_contact, search_contacts, remove_contacts_from_list, create_template_version, add_contact, send_mail - Remove dead data.templates fallback in list_templates (API only ever returns result) - Remove unused UpdateContactParams/UpdateListParams/UpdateTemplateParams dead types; make CreateTemplateParams.generation optional to match actual tool behavior * fix(sendgrid): address Cursor Bugbot findings on pagination and active coercion - Gate listPageToken/templatePageToken remap on operation so a stale token from the other list operation can't override the intended one - Fix active coercion to also treat a real boolean true (from a dynamic <Block.output> reference) as active, not just the dropdown string 'true' * fix(sendgrid): coerce active to int at the tool layer too Per Greptile: the block-level active coercion only covered the UI path. A direct sendgrid_create_template_version tool invocation with a boolean active would still send a raw boolean to SendGrid. Coerce to 0/1 in the tool's own request body so both paths are correct. * fix(sendgrid): always send page_size on list_templates SendGrid's GET /v3/templates requires page_size on every request (no server-side default) — omitting it errors. Default to 20 to match our own documented default when the caller doesn't set one. * fix(sendgrid): explicit false/'false' check for active flag Per Cursor Bugbot: params.active ? 1 : 0 treated any truthy string (including "false") as active. Extracted a toActiveFlag helper that only treats real false or the string 'false' as inactive, everything else (including unset) defaults to active — matches the tool's documented default. * fix(sendgrid): handle numeric 0 in toActiveFlag Per Greptile: the block coerces active to a number (0/1) before calling the tool, but toActiveFlag only checked for false/'false', so the block's inactive selection (0) fell through to the "active" branch. Check against an explicit inactive-values set covering the boolean, string, and numeric forms. * fix(sendgrid): nest add_contact custom fields under custom_fields Pre-existing bug (predates this PR): custom fields were merged onto the contact object as top-level sibling keys via safeAssign/Object.assign, but SendGrid's PUT /v3/marketing/contacts requires them nested under a custom_fields object. SendGrid silently drops unrecognized top-level keys, so the documented customFields param never actually reached SendGrid. Caught during a final adversarial re-verification pass before merge. * fix(sendgrid): document consistent page_size requirement for list_templates pagination Per Cursor Bugbot: list_templates always defaults page_size to 20 when unset (required by SendGrid), so a follow-up pageToken-only call after a first call with a larger pageSize would silently shrink to 20 and desync page boundaries. This is inherent to a stateless tool call (SendGrid requires page_size on every request, and the tool has no way to remember the prior call's value), so clarify via param description and UI placeholder that callers must repeat the same pageSize across paginated calls. * chore(api-validation): bump stale route-count ratchet baseline 883->884 Unrelated to the SendGrid work in this branch. staging's own HEAD already has 884 compliant Zod-backed API routes (0 non-Zod), but this ratchet baseline was never bumped when that route landed, so any PR rebasing onto current staging fails check:api-validation:strict with "route count increased from 883 to 884". All routes remain fully Zod-backed; this is a mechanical counter update, not a policy change. * fix(sendgrid): dedupe active coercion between block and tool Per Cursor Bugbot: the block's pre-coercion only recognized the dropdown string 'true' or boolean true as active, so a dynamic reference producing numeric 1 or string '1' fell through to 0 and silently created an inactive template version. Exported the tool's toActiveFlag and reused it in the block instead of duplicating the inactive-value logic, so both layers can no longer drift out of sync.
A workspace to build, deploy and manage AI agents and workflows.
Quickstart
Cloud-hosted: sim.ai
Self-hosted
npx simstudio
Docker must be installed and running. Use -p, --port <port> to run Sim on a different port, or --no-pull to skip pulling the latest Docker images.
Capabilities
- Connect 1,000+ integrations and every major LLM
- Add Slack, Notion, HubSpot, Salesforce, databases, and more
- Build agents visually, conversationally, or with code
- Ingest files, knowledge bases, and structured table data
- Monitor runs, logs, schedules, and workflow activity
One workspace, every surface
Chat and workflows are just the start — tables, files, knowledge, and scheduled tasks all live in the same workspace.
Tables — a database, built in |
Files — one store for your team and every agent |
Knowledge — your agents' memory |
Scheduled tasks — runs on your schedule |
Self-hosting
Docker Compose
git clone https://github.com/simstudioai/sim.git && cd sim
docker compose -f docker-compose.prod.yml up -d
Sim also supports local models via Ollama and vLLM. See the Docker self-hosting docs for setup details.
Manual Setup
Requirements: Bun, Node.js v20+, PostgreSQL 12+ with pgvector
- Clone and install:
git clone https://github.com/simstudioai/sim.git
cd sim
bun install
bun run prepare # Set up pre-commit hooks
- Set up PostgreSQL with pgvector:
docker run --name simstudio-db -e POSTGRES_PASSWORD=your_password -e POSTGRES_DB=simstudio -p 5432:5432 -d pgvector/pgvector:pg17
Or install manually via the pgvector guide.
- Configure environment:
cp apps/sim/.env.example apps/sim/.env
# Create your secrets
perl -i -pe "s/your_encryption_key/$(openssl rand -hex 32)/" apps/sim/.env
perl -i -pe "s/your_internal_api_secret/$(openssl rand -hex 32)/" apps/sim/.env
perl -i -pe "s/your_api_encryption_key/$(openssl rand -hex 32)/" apps/sim/.env
# DB configs for migration
cp packages/db/.env.example packages/db/.env
# Edit both .env files to set DATABASE_URL="postgresql://postgres:your_password@localhost:5432/simstudio"
- Run migrations:
cd packages/db && bun run db:migrate
- Start development servers:
bun run dev:full # Starts Next.js app and realtime socket server
Or run separately: bun run dev (Next.js) and cd apps/sim && bun run dev:sockets (realtime).
Chat API Keys
Chat is a Sim-managed service. To use Chat on a self-hosted instance:
- Go to https://sim.ai → Settings → Chat keys and generate a Chat API key
- Set
COPILOT_API_KEYenvironment variable in your self-hosted apps/sim/.env file to that value
Environment Variables
See the environment variables reference for the full list, or apps/sim/.env.example for defaults.
Tech Stack
Next.js · Bun · PostgreSQL · Drizzle · Better Auth · Tailwind — and the rest of the stack
- Framework: Next.js (App Router)
- Runtime: Bun
- Database: PostgreSQL with Drizzle ORM
- Authentication: Better Auth
- Schema Validation: Zod
- UI: Shadcn, Tailwind CSS
- Streaming Markdown: Streamdown
- State Management: Zustand, TanStack Query
- Flow Editor: ReactFlow
- Docs: Fumadocs
- Monorepo: Turborepo
- Realtime: Socket.io
- Background Jobs: Trigger.dev
- Remote Code Execution: E2B
- Isolated Code Execution: isolated-vm
Contributing
We welcome contributions! Please see our Contributing Guide for details.
License
This project is licensed under the Apache License 2.0 - see the LICENSE file for details.





