Skip navigation and go to main content

TrolyPage

An autonomous Marketing & Sales operating system for businesses that want AI-powered omni-channel growth.

Contact Info

Follow Us

Every shipped improvement, grouped by day and product.

From AI agent workflows and Live Chat Widget to iOS, Android, Browser Extensions, TrolyMail, and TrolyWork—tracked daily directly from shipped code.

Updates shown: 9 · Days: 9 Matching releases: 9
Showing filtered releases
Latest
Sunday · 2026 1 updates
Security TrolyReddit TrolyIG TrolyThreads Website Chat Mobile App TrolyPage

Digital pre-orders, Telegram delivery polish, and multi-tenant security hardening

Added digital pre-orders, polished Telegram storefront messaging, and strengthened authentication, webhook verification, uploads, and workspace isolation.

Key changes

  • Enabled digital product pre-orders alongside clearer storefront delivery notifications.
  • Hardened tenant isolation, authentication boundaries, webhook signatures, and upload checks across web and mobile.
  • Refined localized interfaces across TrolyReddit, TrolyIG, and TrolyThreads.
Multi-language automation workflows across TrolyPage extensions
Friday · 2026 1 updates
Security TrolyPage

Resource-bound OAuth grants and scoped API credentials for MCP

Hardened MCP connectivity with resource-scoped OAuth grants and revocation, outbound connection safeguards, granular scoped API keys, and guided onboarding.

Key changes

  • Added resource-bound OAuth authorization and revocation with strict outbound transport validation.
  • Introduced scoped API keys, approval status polling, and streamlined MCP client onboarding.
  • Preserved indeterminate token expiration handling alongside dedicated OAuth boundary protections.
Thursday · 2026 1 updates
Security TrolyPage

Hardened dialog focus lifecycle, overlay z-index stacking, and accessibility

Hardens modal overlay focus management, z-index layering, and command palette typography across key workspace dialogs.

Key changes

  • Hardened focus trap lifecycle and restoration across channel studio, tag rules, and agent refinement dialogs.
  • Fixed overlay z-index stacking conflicts and refined command palette accessibility.
Tuesday · 2026 1 updates
Security TrolyPage

AI Reply Safety Gate, Multi-Stage Knowledge Fact Extraction, and Soft-Delete Catalog Protection

Added a pre-delivery AI reply safety gate and PII controls per agent, automated multi-stage knowledge fact extraction and verification, and soft-delete safeguards for products.

Key changes

  • Configure content moderation and personal-data safety gates inside Agent Builder before AI messages reach customers.
  • Extract and verify structured knowledge facts in multiple stages for more accurate product and policy answers.
  • Protected product catalogs with soft deletion while streamlining training package imports in the UI.
Monday · 2026 1 updates
Security TrolyPage

AI Safety Policy Guardrails, Trusted Product Media Verification, and Approval Queues

Added a platform-wide AI safety engine with content moderation, verified product media trust checks, outcome scenarios, and human-in-the-loop Agent Approval continuation flows.

Key changes

  • Protect every AI turn with centralized safety policies, content moderation, and runtime guardrails.
  • Verify product media trustworthiness before AI agents compose carousel cards or order drafts.
  • Review pending agent actions in Agent Approvals and automatically resume customer follow-up once approved.
Monday · 2026 1 updates
Security TrolyPage

Zalo Personal Durable Delivery Stack, Outbox Deduplication, and Tenant Isolation

Built a crash-safe webhook spool, recoverable inbox, and outbound outbox for Zalo Personal to guarantee zero dropped or duplicate messages across AI and staff sends, paired with tenant isolation, SSRF guards, proxy credential redaction, and live health kill-switches.

Key changes

  • Persistent webhook spool with graceful shutdown drain, claim-first inbox recovery, and outbox deduplication for AI and operator replies
  • Strict tenant-scoped identity and authorization checks preventing cross-workspace resends or stale reply replays
  • Hardened outbound security: blocks SSRF on remote media downloads, disables unsafe proxy fallback, and redacts credentials
  • Added safety kill-switches, dead-letter recovery tooling, end-to-end trace correlation, and scheduled connection health alerts
Thursday · 2026 1 updates
Security TrolyPage Mobile App

Platform Security Hardening: Fail-Closed Webhooks, Tenant Guards, SSRF & XSS

Rolled out broad security hardening across web and mobile API endpoints, enforcing fail-closed webhook verification, mass-assignment and cross-tenant access guards, SSRF URL validation, and XSS sanitization.

Key changes

  • Enforced fail-closed webhook verification to reject unsigned or invalid external callbacks across channels
  • Strengthened mass-assignment protection and ownership checks across web controllers and Mobile API profile/auth routes
  • Added strict SSRF protections on extension downloads and hardened output sanitization against XSS vectors
Saturday · 2026 1 updates
Security Mobile App TrolyPage

Mobile API Multi-Workspace Isolation and Per-User Idempotency Hardening

Introduced request-scoped account context resolution and user-scoped idempotency keys across Mobile API endpoints, keeping conversations, assignments, and handoffs strictly isolated across workspaces.

Key changes

  • Request-scoped workspace resolution across mobile conversation, assignment, SLA check, and team workflows
  • Scoped mobile idempotency keys per user so retried actions remain isolated and safe across multi-account teams
Thursday · 2026 1 updates
Security TrolyPage Website Chat

MCP security hardening, Zalo OA reply fix, and Website Chat updates

Strengthened multi-tenant isolation, authentication, and rate limiting on MCP endpoints while fixing Zalo OA webhook replies, enforcing WhatsApp's 24-hour window, refining round-robin routing, and upgrading WebChat visitor avatars.

Key changes

  • Enforces strict tenant isolation, token authentication, and rate limits across MCP server endpoints.
  • Resolves Zalo OA webhook reply issues and adds WhatsApp 24-hour messaging window validation.
  • Improves fair round-robin agent assignment, Facebook avatar caching, and WebChat email avatars with pre-chat context.

STAY IN THE LOOP

Ship faster customer operations with every ecosystem update

Subscribe via RSS, integrate the public JSON API into your workspace workflows, or open TrolyPage to try the latest capabilities.