KRASPER THOT
Krasper Knowledge Hub
Enterprise knowledge management platform that goes far beyond classic documentation. Teams create and review technical documents, import code repositories (GitHub, GitLab, Azure DevOps, Bitbucket, Gitea) with AI-powered auto-documentation, run security scans (SAST), and analyze legacy code (RPG/COBOL) with graph-based dependency visualization.
Why Now — Knowledge Challenges
47%
of organizational knowledge is lost when employees leave
Industry average
8.8h
per week spent by knowledge workers searching for information
6
VCS providers supported — GitHub, GitLab, Azure DevOps, Bitbucket Cloud, Bitbucket Server, Gitea
12+
Bounded Contexts in fully DDD/Hexagonal architecture
Capabilities
Core Features
TipTap Rich-Text Editor
Markdown support, code blocks with syntax highlighting (lowlight), tables, Mermaid diagrams. Side-by-side diff viewer for version comparison.
Review & Approval Workflow
Submit, Approve, Reject, Request Changes. 4-eyes principle (authors cannot review own docs). Complete audit trail per review decision.
Immutable Version History
Every change creates a new version with change description. ETag-based optimistic locking, rollback capability, and complete change history per document.
MeiliSearch Full-Text Search
Blazing-fast search with autocomplete, faceted filtering by categories/tags/author/status, instant results while typing, and dedicated UI view analysis index.
Multi-VCS Repository Import
GitHub, GitLab, Azure DevOps, Bitbucket Cloud, Bitbucket Server, Gitea. Factory pattern with auto-detection, AI-powered documentation generation per file, duplicate detection, stuck import recovery.
Cross-Repository API Mapping
Project-based organization with role assignment (Frontend/Backend/Fullstack/Shared). AST-based extraction, automatic matching engine, AI-assisted confidence scoring, Mermaid graph visualization, coverage dashboard.
Security Scanning (SAST)
Hybrid analysis: Semgrep + Bandit combined with LLM-enriched findings. Vulnerability categorization by severity and CWE, risk scoring, AI-generated remediation suggestions, dashboard with trends.
Secrets Detection
AI-based detection of credentials, API keys, and tokens in imported code. Confidence scoring per finding and resolution tracking.
RPG/COBOL Legacy Analysis
Multi-language parser (RPG, SQLRPGLE, CL, COBOL, Java, Python, Go, C#, TypeScript). Neo4j graph database for dependency graphs, Cytoscape visualization, impact analysis, cycle detection.
UI Guide Auto-Generation
Automatic UI documentation from imported frontend code. Framework detection (Vue, React, Angular, Svelte, Astro, PHP, Django). 3-phase pipeline: Discovery, Per-View Analysis, Synthesis. Versioning with changelog.
Multi-Channel Notifications
In-App, Email, Slack, Webhook. Event-driven alerts for reviews, comments, imports, scans. Per-user preferences, watch subscriptions, digest mode.
Document Export
PDF (WeasyPrint), DOCX (python-docx), HTML, Markdown. Customizable Jinja2 templates, bulk export, export history tracking.
Architecture
Domain-Driven. Self-Hosted. Enterprise-Grade.
Built on hexagonal architecture with 12+ bounded contexts, 85 use cases, and 80+ ports — zero dependency rule violations.
DDD Backend
12+ bounded contexts, 85 use cases with single-execute pattern, 80+ protocol-based ports, 25+ SQLAlchemy repositories, 26 in-memory fakes for testing, 416 architecture fitness tests.
DDD Frontend
16 domain modules, ~130 Vue components, 14 DDD stores + 12 legacy stores, architecture fitness tests with import rules, readonly domain types, strict TypeScript.
Infrastructure
11+ Docker services, 5 Celery worker queues with beat scheduler, PostgreSQL 18, Redis 7, MeiliSearch 1.12, Neo4j 5. Nginx with TLS 1.2/1.3, Let's Encrypt.
Self-Hosted. Fully Encrypted. Full Control.
Thot runs entirely on your infrastructure. Argon2id password hashing, TOTP/2FA, Azure Entra ID SSO, CSRF protection, SSRF prevention, Fernet encryption for all credentials, and comprehensive rate limiting. Your documentation and code analysis never leave your perimeter.
AI Pipelines
Five AI Pipelines. Fully Async.
All LLM calls run asynchronously via dedicated Celery workers with configurable timeouts and race-condition protection.
Repository Import Docs
Concurrency: 2Queue: repository_imports
AI-generated documentation per imported file
Security Scanning
Concurrency: 3Queue: security_scanning
Hybrid Semgrep/Bandit + LLM enrichment
Secrets Detection
Concurrency: 3Queue: secrets_scanning
LLM-based credential detection in batches
UI Guide Generation
Concurrency: 1Queue: ui_guide_generation
3-phase pipeline: Discovery, Analysis, Synthesis
Cross-Repo Mapping
Concurrency: 1Queue: cross_repo_mapping
LLM-assisted frontend-backend matching
Target Audience
Built For Technical Teams
Engineering Teams
Technical documentation management and quality assurance
Security Teams
SAST scans and secrets detection on imported repositories
Legacy Modernization
RPG/COBOL codebase analysis and dependency understanding
Multi-Repo Projects
Cross-repository API mappings for microservice architectures
DevOps Teams
Webhook-based automation of documentation and scan workflows
Technical Departments
Knowledge management for product and system documentation
Where Thot fits in this landscape
Confluence is the de-facto standard but trending cloud-only with no review workflow or code analysis. Notion is modern but cloud-only with no SAST. BookStack is open-source self-hosted but lacks review workflows and AI features. Outline has limited versioning and no security scans. GitBook is developer-focused but has no SAST integration or legacy code support. Thot combines structured review workflows, AI-powered code documentation, integrated SAST with LLM enrichment, RPG/COBOL graph analysis, cross-repo API mapping, multi-VCS support, and immutable audit trails — all self-hosted with DDD architecture.
Transparent Pricing
Knowledge Management at Every Scale
Thot Team
$2,000/mo
$24,000 ARR
Thot Individual
Custom
Custom pricing
Ready to Centralize Your Knowledge?
Start with a 90-day pilot. We import your repositories, configure your workflows, and deliver measurable documentation quality improvements — or you walk away.