Echo-Synch

Security & trust

Trust Center

We treat your IT request data with the same discipline we expect of an SLA queue. Transparency is our default setting.

Infrastructure & architecture

Minimal-footprint processing

Echo-Synch reads message content only from channels you explicitly add the bot to. Original IT request text is auto-scrubbed at each workspace's configured retention window (default 30 days). Triage metadata — labels, priorities, AI summaries, SLA timer events — is retained while the workspace is installed and hard-deleted within one hour of uninstall.

Encryption

All data is encrypted in transit via TLS 1.2+ (1.3 preferred) and at rest using AES-256. Slack bot tokens are encrypted with AWS KMS before being written to the database. Application secrets live in AWS SSM Parameter Store with at-rest encryption.

Cloud security

Hosted on AWS in eu-west-1 (Dublin, Ireland) using serverless architecture (Lambda + Aurora Serverless v2 + SQS) for strict isolation between requests. All processing and storage happens in the EU.

Least privilege

We request the minimum Slack OAuth scopes needed to operate. Each scope has a justification documented at its call site in our codebase (see privacy for the list and rationale). Workspace admins can audit and revoke access at any time via Slack app management.

Compliance

Sub-processors

Echo-Synch uses the third-party processors below, each contractually bound to process your data only on our instructions. The complete list with locations and purposes is in Annex 1 of the DPA.

Sub-processorPurposeLocation
Amazon Web Services (AWS)Hosting & infrastructureIreland (eu-west-1)
Google (Vertex AI / Gemini)Default AI providerBelgium (EU)
Google Gemini under BYOMEnterprise BYOM — runs in the Customer's own GCP projectCustomer-selected region
LemonSqueezyPayment processing & merchant of recordUSA (DPF-certified)
CloudflareDNS, CDN, email routingGlobal edge / EU primary
ResendTransactional emailUSA (DPF-certified)
SentryError monitoring (PII attachment disabled)USA (DPF-certified)

We will provide reasonable prior notice (via this page and email to the installer) of any addition or replacement of sub-processors.

Bring Your Own Model (BYOM)

Complete data isolation

With BYOM, message content is sent directly to your own Google Cloud project — never to Echo-Synch's. Your AI credentials, your infrastructure, your control. Echo-Synch is not in the AI data path.

Data sovereignty

BYOM lets you control exactly where AI processing happens. Choose any GCP region your organisation has approved — your IT request content never leaves the regulatory boundary you define.

No model training

Whether you use our default Vertex AI provider or BYOM, your data is never used to train AI models. Processing is ephemeral and stateless. Vertex AI's default API contract explicitly excludes training use of customer prompts.

Operational practices

Data residency

Primary infrastructure runs in AWS Ireland (eu-west-1), Vertex AI runs in Google Cloud Belgium. Backups and snapshots remain in Ireland. US-based subprocessors (LemonSqueezy, Resend, Sentry) operate under the EU–US Data Privacy Framework with Standard Contractual Clauses incorporated.

Questions about security?

Enterprise security questionnaires answered within 5 business days. Email security@echo-synch.com.