Product Updates
Changelog
All notable changes to Protectron are documented here.
Version Format
We use Semantic Versioning:
- Major (1.x.x) — Breaking changes
- Minor (x.1.x) — New features, backward compatible
- Patch (x.x.1) — Bug fixes, backward compatible
[1.0.0] — 2025-12-16
Official V1 Release
We're excited to announce the official V1 release of Protectron — the complete EU AI Act compliance platform for AI systems and agents.
Features
Dashboard
- Risk Classification Engine — Automatically classify AI systems by EU AI Act risk level
- Requirement Tracking — Track all 113 high-risk AI requirements with progress monitoring
- Compliance Score — Real-time compliance score calculation across all your AI systems
- Multi-System Management — Manage unlimited AI systems from a single dashboard
- Team Collaboration — Invite team members with role-based access
Document Generation
- AI-Powered Generation — Generate EU AI Act documentation using AI
Document Types:
- • Technical Documentation (Article 11)
- • Risk Management System (Article 9)
- • Data Governance Documentation (Article 10)
- • Human Oversight Procedures (Article 14)
- • Transparency Documentation (Article 13)
- • AI Governance Policy
- • AI Ethics Policy
- • Incident Response Plan
- Export Formats — PDF, DOCX, and Markdown export
- Version History — Track document changes over time
SDK & Audit Trail
- Python SDK — pip install protectron
- TypeScript SDK — npm install @protectron/sdk
- LangChain Integration — Native callback support
- CrewAI Integration — Multi-agent crew logging
- AutoGen Integration — Microsoft AutoGen conversation tracking
Automatic Event Logging:
- • LLM calls (prompts, completions, tokens, latency)
- • Tool/function invocations
- • Agent actions and decisions
- • Agent delegation and collaboration
- • Human oversight interventions
- • Errors and exceptions
- PII Redaction — Automatic detection and redaction of sensitive data
- Trace Grouping — Group related events into logical traces
Evidence Management
- Document Upload — Upload existing compliance evidence
- Requirement Linking — Link evidence to specific requirements
- Audit Packages — Generate downloadable audit packages
Certification Badges
- Embeddable Badges — Display compliance status on your website
- Real-Time Verification — Visitors can verify with one click
- Multiple Badge Types — Standard, Detailed, Minimal, System-Specific
API
- REST API v1 — Full programmatic access to all features
- Webhooks — Real-time notifications for compliance events
Security
- EU Data Residency — All data stored in AWS EU (Frankfurt)
- Encryption — TLS 1.3 in transit, AES-256 at rest
- SOC 2 Type I — Certified (Type II in progress)
- GDPR Compliant — Full compliance with EU data protection
- SSO Support — SAML 2.0 and OIDC (Enterprise plans)
SDK Packages
| Package | Version | Install |
|---|---|---|
| protectron | 1.0.0 | pip install protectron |
| protectron[langchain] | 1.0.0 | pip install protectron[langchain] |
| protectron[crewai] | 1.0.0 | pip install protectron[crewai] |
| protectron[autogen] | 1.0.0 | pip install protectron[autogen] |
| @protectron/sdk | 1.0.0 | npm install @protectron/sdk |
Documentation
- Quick Start Guide — Get started in 5 minutes
- SDK Overview — Complete SDK documentation
- API Reference — REST API documentation
- LangChain Guide — LangChain integration
- CrewAI Guide — CrewAI integration
Pricing Plans
| Plan | Price | AI Systems | Events/month |
|---|---|---|---|
| Starter | €99/mo | 3 | 10,000 |
| Growth | €299/mo | 10 | 100,000 |
| Scale | €599/mo | 25 | 500,000 |
| Enterprise | Custom | Unlimited | Unlimited |
All plans include a 14-day free trial.
Known Issues
- Large PDF exports (>100 pages) may timeout on Starter plans
- Safari 15.x may show badge rendering issues (fix in progress)
- Webhook retry logic limited to 3 attempts in V1
Coming Soon
Features planned for upcoming releases:
- Slack Integration — Compliance alerts in Slack
- Jira Integration — Sync requirements with Jira tickets
- AutoGen v0.4 Support — Updated AutoGen integration
- Vercel AI SDK — Native TypeScript integration
- Multi-language Documents — Generate docs in DE, FR, ES, IT
- Compliance Templates — Industry-specific templates
- Custom Branding — White-label badges and reports
Pre-Release History
[0.9.0-beta] — 2025-11-15
Beta Release
- • Initial public beta
- • Core dashboard functionality
- • LangChain integration
- • Basic document generation
[0.8.0-alpha] — 2025-10-01
Alpha Release
- • Private alpha for early adopters
- • Risk classification engine
- • Basic event logging
Upgrade Guide
Migrating to V1
If you were using the beta SDK, update your packages:
Python:
pip install --upgrade protectronTypeScript:
npm install @protectron/sdk@latestBreaking Changes from Beta
1. Callback class renamed:
# Beta
from protectron import ProtectronLangChainCallback
# V1
from protectron.langchain import ProtectronCallback2. Configuration options restructured:
# Beta
callback = ProtectronCallback(api_key="...", system="my-agent")
# V1
callback = ProtectronCallback(system_id="my-agent") # API key from env3. Event types standardized:
- • llm_request → llm_call
- • function_call → tool_call
- • agent_step → agent_action
Feedback
We'd love to hear from you:
- Feature Requests: github.com/protectron/sdk/issues
- Bug Reports: github.com/protectron/sdk/issues
- General Feedback: feedback@protectron.ai
Subscribe to Updates
Get notified about new releases:
- GitHub: Watch github.com/protectron/sdk
- Email: Subscribe in your dashboard settings
- Twitter/X: Follow @protectron_ai
Last updated: December 16, 2025