Town

Read and write integrations reach Google, Microsoft and more than fifty business tools: Gmail (read, search, draft, send, label), Google Calendar (create meetings with Meet links, edit events), Docs, Drive, Sheets, Outlook and Microsoft 365 mail and calendar, Slack (bot and send-as-user), and read-write connectors across CRM (Salesforce, HubSpot, Attio, Pipedrive, Copper, Affinity), project tools (Asana, Jira, Linear, ClickUp, Monday.com, Notion, Todoist), finance (QuickBooks Online and Xero create invoices, Brex, Ramp, Stripe), data (Airtable, Supabase, PostHog, Google Analytics, Hex) and meetings (Zoom, Fathom, Fireflies, Granola, Gong).

An admin page covers Microsoft (Entra). Connectors are native plus Pipedream-backed, actions create, update and send, and OAuth tokens are stored, refreshed automatically and revocable per account.

WhatsApp and Telegram are inbound-only. Custom MCP servers can be added too.

Sourcetown.com home; town.com/docs, /docs/safety/security; readread 2026-09-16

Routines can call other routines and mix deterministic code steps with AI steps. On Combining Routines, "a routine can call another routine mid-run": the caller pauses, the called routine returns its result, and a parent must list the child in its Callable Routines setting before invoke_routine can call it.

The page separates AI steps (writing, summarizing, deciding, analyzing) from deterministic steps (filtering, formatting, aggregating, iterating over a list) and walks a four-step example: gather (code), organize (code), write (AI), send (code). Users describe a routine in plain language, set it on a trigger or schedule, and share routines across a team.

Versioning is not described.

Sourcetown.com/docs/routines/custom-routines/concepts/calling-agents, /docs; readread 2026-09-16

The assistant is grounded in the user's connected email, calendar, chats, contacts and tools, queried at run time rather than indexed.

The context page says the Townie draws on "email, calendar, chats, and contacts", and "every integration you add (Slack, Notion, Drive, Linear, GitHub, and more) becomes another source of context"; integration pages describe search-and-read tools (Gmail search, Drive search and read, Box files and metadata, Notion pages and databases, Granola and Fireflies transcripts).

Connected sources are permissioned per account and by routine account scope, and new knowledge enters by connecting a source rather than retraining.

No index or search layer over connected content persists between runs. The Wiki and memories are syntheses about the user, not an index of their content.

Sourcetown.com/docs/features/context-and-memory, /docs integration directory; readread 2026-09-16

Approval modes are configurable at the routine, tool and session level. Every routine has a default mode: autonomous; approval-required, where actions that make changes create approval requests with approve or reject options and "only execute after you approve them"; or read-only. Per-tool permission levels override the routine mode (Auto, Ask each time, Always allow), so labeling can run autonomously while email sending always waits.

Chat sessions run "Ask before any changes" by default, "Always allow safe actions" or "Allow all actions", with Allow once and Always allow per approval. Town also flags routines that combine private data, untrusted input and external sending (a lethal-trifecta warning) when they are saved, and keeps external-send tools on approval under safe mode. Low-risk and high-risk actions can run under different autonomy modes, and the warning explains why a checkpoint fired.

Routing approvals into ticketing systems is not described beyond app and notification delivery.

Sourcetown.com/docs/safety/modes-approvals, /docs/safety/security; readread 2026-09-16

An independently audited SOC 2 Type 2 sits beside scope controls customers set themselves. The home page says "SOC II Type 2 compliant: Town's security controls are independently audited", beside links to a Security Portal and Subprocessors.

Routines operate within strict account boundaries with configurable scope (the trigger account only by default, all accounts, or specific accounts), tools can only reach accounts connected through OAuth with revocable access, and optional domain allowlists and recipient limits apply to email.

Prompt validation blocks instruction-override and exfiltration patterns, and lethal-trifecta detection flags routines that combine private data, untrusted input and external sending. Data is not used to train AI models unless the user opts in, and is encrypted in transit and at rest.

People sign in with Google or Microsoft accounts, including Microsoft Entra ID, and Teams plans add domain capture. An Entra admin can grant consent for the whole tenant and then either let every user in the tenant sign in or restrict access to chosen users and security groups, and Google Workspace has its own admin guide. SAML single sign on and SCIM provisioning for team admins are not stated.

SOC 2 Type 1 and Type 2 reports, the latest penetration test and a CASA Tier 2 assessment are available under NDA through trust.town.com. Traffic uses TLS 1.2 or 1.3, data at rest uses AES 256 GCM with extra encryption on connected account tokens, Enterprise agreements can enforce no training for every user, and no Town team member has production database access.

SourceTown, town.com/docs/admin-setup/microsoft-entra, /docs/safety/security and town.com/trustread 2026-10-05

Every action is logged with its trigger, reasoning and result, and a separate security event log sits beside it. Under action logging, every action is logged with what (the action taken), when, why ("the routine's reasoning"), trigger (what initiated it) and result (success or failure), for auditability, debugging and accountability.

Users open a routine, view the session history, and review actions taken and their reasoning. A separate security event log records blocked prompt injections, suspicious prompts, blocked tool outputs and unauthorized recipients, and custom tool results show up in the run history. That is run-level visibility of the steps taken and the tools called, with an audit log separate from runtime traces.

Export to monitoring or SIEM tools and retention by plan are not documented.

Sourcetown.com/docs/safety/security, /docs/integrations/mcp-servers; readread 2026-09-16

Memory comes with a stated scope, review and deletion, plus a wiki about the user that refreshes nightly. The Townie remembers "who you are, your preferences" and "contacts & people notes"; memories are global (applying to everything, across all routines) or routine-specific (applying to a single routine), and a run automatically receives all global memories plus those for that routine.

Memory "carries over" between Slack, email, the app and the Mac app, and users view and delete memories in Settings > Profile or by asking ("Forget the preference about Friday afternoons"), and can "add, change, or remove anything". A Wiki synthesizing profile, preferences, goals and projects is "refreshed every night", retiring what is stale, and is editable line by line, with Recent Updates showing what changed. Memories last until the user removes them, with nightly retirement of stale entries, and can be deleted without deleting any email, event or document.

What should never enter long term memory is not stated. Connected email, calendar and files are knowledge sources rather than memory.

Sourcetown.com/docs/features/context-and-memory, /docs/features/wiki; readread 2026-09-16

Hosting is in the US, run by the vendor. The home page's security section says Town is "US-hosted on Convex and AWS"; no region choice, self-hosting or customer-environment option is offered, and the Teams and Enterprise plans describe billing and credits, not deployment.

Sourcetown.com home; town.com/docs/billing/plans; readread 2026-09-16

Ready made routines sit in a browsable catalog, each a complete job a user enables.

The home page catalog filters by Inbox, Calendar, Meeting Prep, Stay Informed and Relationships, with "Browse all Routines"; entries include Auto-inbox (watches new email, labels, drafts replies, helps schedule), Decline Cold Outreach, Schedule Optimizer, Meeting Briefing, Contact Research Dossier, Morning Briefing, Competitive Intel Briefing, Newsletter Digest, GitHub Reports, Relationship Reconnect, Deal Spotter and New User Research.

The docs carry a Stock Routines index with per-routine pages and routine sharing and discovery, so teams install and reuse shared routines. Each routine runs whole without the others, and routines are edited in plain language ("change my morning briefing to include...").

The routines are personal productivity patterns rather than vertical packs, and customer references are not tied to them.

Sourcetown.com home; town.com/docs; readread 2026-09-16

Work reaches Town's routines without a person asking, on email events, schedules and calendar events, across many channels. Each routine can have one or more triggers and runs whenever any fires: incoming email ("runs every time a new email arrives"), a smart email trigger that judges a plain-language condition with sender and subject filters, outgoing email, email to the assistant, schedule and manual runs; meeting-start (calendar) triggers come on Pro and above.

A new email or a schedule firing starts a routine with nobody asking. Triggers are event-driven, scheduled and conversational, several can feed one routine, and the smart trigger "runs anyway" if its condition cannot be judged. The Townie is reachable by email, Slack, text messaging, WhatsApp and Telegram (inbound-only), the web app, iOS and macOS.

Event and calendar triggers need Pro or a paid team plan, and Starter is schedule only, at most once a day. An authenticated webhook endpoint can also start a routine.

Sourcetown.com/docs/routines/custom-routines/triggers, /docs/billing/plans; readread 2026-09-16

Routing between models happens on the vendor's side, with no models disclosed and no choice offered. The pricing page lists "priority model routing" as a feature of the Power, Power Plus and Team Plus plans, and among what pauses on cancellation; no page names a model or provider or lets the customer choose (the Cursor integration's models are Cursor's). The home page's no training statement covers "our providers" without naming them.

Sourcetown.com/pricing; readread 2026-09-16

Outside systems can reach Town two ways, but no REST API, SDK or served MCP for the platform itself is published.

Webhook triggers: an external system "starts a routine by sending an authenticated HTTP request to a routine-specific URL", with a per-routine bearer secret and rotation, JSON or text payloads, an idempotency header, rate limits and documented 202/400/404/413/415/429 responses, usable from Zapier or "your own app or backend".

Custom integrations (MCP): users add MCP servers by URL, Town runs OAuth or dynamic client registration, and the tools are enabled per routine, so Town consumes MCP as a client.

A trigger endpoint returns only an acceptance, and acting as an MCP client shows what Town can call, not whether Town itself is callable. The api.town.com host resolves to the Convex backend banner, not a developer surface.

Sourcetown.com/docs/routines/stock-routines/webhooks, /docs/integrations/mcp-servers; api.town.com; readread 2026-09-16

A new or changed routine can be tried against real input, with its attempted actions reviewable, before it is allowed to act. Read-only mode runs a routine that can read and analyze but blocks every change, and its documented use is "testing new configurations"; the manual trigger is documented for "testing new routines".

Each run's session history shows the actions the routine took and its reasoning, and read-only runs still send their results to the user. A read-only run reviewed before the routine is switched to approval-required or autonomous reports on agent behavior before release. When a smart email trigger cannot judge its condition, the routine runs anyway.

There are no fixtures or datasets, scored test cases or output quality scoring over time, and testing is a documented use of a mode rather than a dedicated harness.

Sourcetown.com/docs/safety/modes-approvals, /docs/routines/custom-routines/triggers; readread 2026-09-16

No browser, desktop or computer control capability is documented. Actions run through named integrations with read-write tools, a run_code sandbox with optional internet access, and email, Slack and messaging channels. Town Desktop is described as a macOS app for "native desktop workflows and notifications", a client for reaching the assistant, not an agent operating the screen. Terminal and shell access are not computer use.

Sourcetown.com/docs, /docs/safety/security; town.com home; readread 2026-09-16