b674d85583
Fix SafeCommand to allow shell scripts with redirects and command chaining
...
- Add shell_script_arg() method for bash/sh/cmd -c scripts
- Allow > < redirects in shell scripts (blocked in regular args)
- Allow && || command chaining in shell scripts
- Update safe_sh_command functions to use shell_script_arg
- Update run_commands, start, and LLM server commands
- Block dangerous patterns: backticks, path traversal
- Fix struct field mismatches and type errors
2026-01-08 23:50:38 -03:00
5919aa6bf0
Add video module, RBAC, security features, billing, contacts, dashboards, learn, social, and multiple new modules
...
Major additions:
- Video editing engine with AI features (transcription, captions, TTS, scene detection)
- RBAC middleware and organization management
- Security enhancements (MFA, passkey, DLP, encryption, audit)
- Billing and subscription management
- Contacts management
- Dashboards module
- Learn/LMS module
- Social features
- Compliance (SOC2, SOP middleware, vulnerability scanner)
- New migrations for RBAC, learn, and video tables
2026-01-08 13:16:17 -03:00
a5dee11002
Security audit: Remove all production .unwrap()/.expect(), add SafeCommand, ErrorSanitizer
...
- Phase 1 Critical: All 115 .unwrap() verified in test code only
- Phase 1 Critical: All runtime .expect() converted to proper error handling
- Phase 2 H1: Antivirus commands now use SafeCommand (added which/where to whitelist)
- Phase 2 H2: db_api.rs error responses use log_and_sanitize()
- Phase 2 H5: Removed duplicate sanitize_identifier (re-exports from sql_guard)
32 files modified for security hardening.
Moon deployment criteria: 10/10 met
2025-12-28 21:26:08 -03:00
c67aaa677a
feat(security): Complete security infrastructure implementation
...
SECURITY MODULES ADDED:
- security/auth.rs: Full RBAC with roles (Anonymous, User, Moderator, Admin, SuperAdmin, Service, Bot, BotOwner, BotOperator, BotViewer) and permissions
- security/cors.rs: Hardened CORS (no wildcard in production, env-based config)
- security/panic_handler.rs: Panic catching middleware with safe 500 responses
- security/path_guard.rs: Path traversal protection, null byte prevention
- security/request_id.rs: UUID request tracking with correlation IDs
- security/error_sanitizer.rs: Sensitive data redaction from responses
- security/zitadel_auth.rs: Zitadel token introspection and role mapping
- security/sql_guard.rs: SQL injection prevention with table whitelist
- security/command_guard.rs: Command injection prevention
- security/secrets.rs: Zeroizing secret management
- security/validation.rs: Input validation utilities
- security/rate_limiter.rs: Rate limiting with governor crate
- security/headers.rs: Security headers (CSP, HSTS, X-Frame-Options)
MAIN.RS UPDATES:
- Replaced tower_http::cors::Any with hardened create_cors_layer()
- Added panic handler middleware
- Added request ID tracking middleware
- Set global panic hook
SECURITY STATUS:
- 0 unwrap() in production code
- 0 panic! in production code
- 0 unsafe blocks
- cargo audit: PASS (no vulnerabilities)
- Estimated completion: ~98%
Remaining: Wire auth middleware to handlers, audit logs for sensitive data
2025-12-28 19:29:18 -03:00
