Everything 8200.dev does

Every AI agent and service account on a scope-vs-trust risk matrix, with its exposure. You set the access policy each agent is allowed; when one exceeds it you get a finding with the exact fix — and, once you grant write access, an admin can revoke or reduce that access from 8200.dev.

An interactive map of who shares what with whom outside your org, with external hubs highlighted. Each risky flow gets a plain-language verdict — allow, flag or block — against the destinations and sensitivity you allow; a block is enforced only when prevention is switched on for that connector, and shown as a simulation until then.

For every detection rule, decide what happens: override its severity, switch it off, route its alerts, and choose its tier — Detect & Alert, Recommend & Guide or Auto-remediate. One global kill switch returns everything to simulate-only.

Proactive watch-and-act rules that alert or queue a fix the moment a new finding matches — scan-based and honest.

Step-by-step fix guidance for every finding. When you arm it, 8200.dev revokes or reduces access itself: public links, external shares and unused or orphaned API keys can be removed unattended; downgrades always wait for a person's click. Every action is audit-logged with the state before and after.

One view answering “what AI is running in your org, and is it governed?” — shadow AI discovered through OAuth grants (DeepSeek, Kimi, Grok, Perplexity), agentic platforms like Manus on elevated watch, vendor AI-training posture, and governance findings from the connected AI vendors (Mistral AI, OpenAI, ChatGPT Enterprise, Anthropic, Cursor). Point-in-time posture from read-only admin-API and OAuth metadata — no traffic interception, no prompt inspection.

A complete, continuously-updated inventory of every AI agent and service account that can reach your data. You cannot govern — or prove you governed — what you cannot see.

Flags when a connected platform is set to use your organization’s content or metadata to train third-party AI models, so it’s a decision you make on purpose, not a default you never saw.

Manually register every AI agent on a platform we don't connect to yet — WhatsApp, Telegram, ManyChat, Tidio, custom and voice bots. Classify each agent's data access, get an automatic risk score and findings, and export the audit-ready proof of inventory. Discovered + registered = your complete inventory.

Flag the OAuth apps built with AI app builders — Lovable, Base44, Bolt.new, Cursor and similar — see exactly what data each can reach, and get findings the moment one touches customer PII or holds broad access with no retention policy. You deployed it; you're accountable for it.

Find the third-party and AI OAuth grants already connected to your Workspace, so an unsanctioned tool is caught before it becomes a liability.

An immutable, exportable record of access decisions (CSV / JSON / CEF / SIEM) — the audit-ready evidence a reviewer or auditor expects.

Generate the documentation auditors request, mapped to SOC 2, ISO 27001, and GDPR controls. Maps to frameworks; it never claims certification.

Connect Intercom and discover the Fin AI Agent and every custom bot — their knowledge sources, whether they answer on sensitive topics, the external data they reach, and whether a human is in the loop. You are accountable for what a customer-facing AI says; govern it before it speaks.

Connect Zendesk and surface its AI bots and agent Copilot — sensitive-field and PII access, escalation policy, third-party AI apps — alongside account security (SSO, 2FA enforcement, public ticket sharing, API-token age and custom-role delete rights).

Public links, external sharing, stale access, over-permissioning, owner sprawl, AI-agent reach and more — every scan, read-only.

Inventory every Shared Drive, its members and external access, restriction settings and orphaned-manager risks.

Discover which marketplace and OAuth apps users granted access to your Workspace, with a risk score per app.

Flag files whose names or locations match credential, financial, HR and PII patterns — tunable per organization.

Build your own checks in an 18-field rule builder — no code — and fold them into every scan alongside the built-ins.

Per-user baselines plus seven anomaly types — sharing spikes, bulk access, dormant reactivation, OAuth bursts.

A per-user security score with a leaderboard and badges, so good data hygiene becomes visible and rewarded.

One chronological feed of findings, scans, audit events and OAuth grants — the unified story of your security posture.

See how your posture score compares to anonymized peers in your industry and company-size band.

Track mean-time-to-remediate, SLA attainment and team activity over time — see how fast risks get fixed and where the bottlenecks are.

Every external share in one place with an approval workflow, trusted-domain trust list and a sharing report PDF.

Scan findings grouped into one alert per category, routed by severity, with quiet hours and noise-killing suppression.

A full incident lifecycle (Open → Investigating → Contained → Resolved → Closed) with an exportable response PDF.

Map findings to SOC 2, ISO 27001, GDPR, HIPAA and NIST CSF controls — Satisfied / Partial / Gap, no spreadsheets.

One audit-ready ZIP per framework — a folder per control with the product data that proves it. Secret-free.

A single 0–100 score, decomposed by category and rule, with the top improvements and quick wins ranked.

A board-ready, AI-written summary PDF over fact-only data — risks, trend, peer comparison and recommended actions.

Opt into a weekly or monthly posture report emailed as a PDF — no one has to remember to run it.

Rich Block Kit finding cards with Acknowledge / Snooze buttons, slash commands and per-org install — no SDK.

Auto-create Jira, Linear, GitHub or Asana tickets for new findings, with two-way status sync that auto-resolves.

HMAC-signed webhooks for every security event, plus Slack/Teams formatting and SIEM (CEF) presets.

Stream the audit trail to your SIEM — CSV / JSON / CEF export and a live event stream for enterprise tooling.

Programmatic access to findings and posture with scoped API keys, Bearer auth and rate limits.

Enterprise single sign-on with Okta, Azure AD, Google and OneLogin — metadata import, JIT provisioning, require-SSO.

Organization-wide multi-factor enforcement with TOTP authenticator apps and one-time backup codes.

Fine-grained roles built from per-feature permission atoms, with templates — beyond Owner / Admin / Viewer.

Restrict access to your trusted CIDR ranges (IPv4 and IPv6), with self-lockout prevention.

See active sessions, set max-duration and idle-timeout policies, and force-logout a user.

A per-organization, append-only trail of who did what and when — the record an auditor asks for.

A grounded, multilingual sales-and-support agent that answers product questions in your language, day or night.

The positive “what do I have” view — users, files and Shared Drives, searchable, filterable and exportable.

13 hand-translated locales plus 20 more translated on the fly, with full right-to-left support.

Manage several organizations under one login and switch between them in a click — built for teams and partners.

Two dozen how-to articles, a searchable help center, and a public status page with uptime monitoring.

A usage dashboard, invoice history and self-serve plan changes — no sales call to upgrade or cancel.