DeepSeek R1 model outputs reasoning_content first, then content.
With n_predict=50, responses were truncated during reasoning phase.
Increased to 512 to allow full reasoning + response.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Change model paths to use ./data/llm/ instead of relative paths from build dir
- Remove cd command when starting llama-server to keep botserver root as cwd
- This fixes model loading when servers are started from different directories
- Both LLM and embedding servers now start successfully
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Add vector_db_health_check() function to verify Qdrant availability
- Add wait loop for vector_db startup in bootstrap (15 seconds)
- Fallback to local LLM when external URL configured but no API key provided
- Prevent external LLM (api.z.ai) usage without authentication key
This fixes the production issues:
- Qdrant vector database not available at https://localhost:6333
- External LLM being used instead of local when no key is configured
- Ensures vector_db is properly started and ready before use
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
The config_manager.get_config() can return Ok("") for empty config values,
which would pass through unwrap_or_else() without using the default.
Added checks after config retrieval to use defaults when config values
are empty strings:
- gpu_layers: "20" (default for GPU layers)
- n_moe: "4" (default for MoE)
- parallel: "1" (default for parallel)
- n_predict: "50" (default for predict)
- n_ctx_size: "32000" (default for context size)
This fixes the error: "error while handling argument --n-gpu-layers: stoi"
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
When no default.gbai/config.csv exists, the system now:
- Sets default llm_server_path to ./botserver-stack/bin/llm/build/bin
- Uses correct relative paths to model files: ../../../../data/llm/
- Uses actual model filenames from 3rdparty.toml
This fixes the issue where LLM/embedding servers couldn't find model files
because the paths were constructed incorrectly.
Model filenames:
- LLM: DeepSeek-R1-Distill-Qwen-1.5B-Q3_K_M.gguf
- Embedding: bge-small-en-v1.5-f32.gguf
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
Changed the default LLM model from glm-4 to deepseek-small to match
the model defined in 3rdparty.toml ([models.deepseek_small]).
This ensures that when no default.gbai/config.csv exists, the system
uses the correct default local model.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
When no default.gbai/config.csv exists or when llm-model/embedding-model
config is empty, the system now uses default local models instead of
skipping server startup.
Changes:
- Default LLM model: glm-4
- Default Embedding model: bge-small-en-v1.5
- Logs when using defaults
This fixes the issue where the "default" bot would fail to load LLM
and Embedding services when no config.csv was present, causing the
error: "not loading embedding neither llm local for default bot"
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Add Redis-based tracking to prevent start.bas from running repeatedly
when clicking suggestion buttons. start.bas now executes only once per
session with a 24-hour expiration on the tracking key.
- Add generic tool executor (ToolExecutor) for parsing and executing
tool calls from any LLM provider. Works with Claude, OpenAI, and
other providers that use standard tool calling formats.
- Update both start.bas execution paths (WebSocket handler and LLM
message handler) to check Redis before executing.
- Fix suggestion duplication by clearing suggestions from Redis after
fetching them.
- Add rate limiter for LLM API calls.
- Improve error handling and logging throughout.
Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>
- Fix ConfigManager to treat 'none', 'null', 'n/a', and empty values as placeholders
and fall back to default bot's configuration instead of using these as literal values
- Fix ConfigManager to detect local file paths (e.g., .gguf, .bin, ../) and fall back
to default bot's model when using remote API, allowing bots to keep local model
config for local LLM server while automatically using remote model for API calls
- Fix get_default_bot() to return the bot actually named 'default' instead of
the first active bot by ID, ensuring consistent fallback behavior
- Add comprehensive debug logging to trace LLM configuration from database to API call
This fixes the issue where bots with incomplete or local LLM configuration would
fail with 401/400 errors when trying to use remote API, instead of automatically
falling back to the default bot's configuration from config.csv.
Closes: #llm-config-fallback
- Add llm to default features in Cargo.toml
- Fix duplicate smart_router module declaration
- Remove unused LLMProvider import and fix unused variable warnings
- Fix move error in enhanced_llm.rs by cloning state separately for each closure
- Improve code formatting and consistency
- shell_script_arg blocks $( and backticks for user input safety
- trusted_shell_script_arg allows these for internal installer scripts
- Internal scripts need shell features like command substitution
- Updated bootstrap, installer, facade, and llm 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
- Add generate_secure_password() for OTP generation during admin bootstrap
- Display admin credentials (username/password) in console on first run
- Save credentials to ~/.gb-setup-credentials file
- Fix Zitadel client to support PAT token authentication
- Replace OAuth2 password grant with Zitadel Session API for login
- Fix get_current_user to fetch user data from Zitadel session
- Return session_id as access_token for proper authentication
- Set email as verified on user creation to skip verification
- Add password grant type to OAuth application config
- Update directory_setup to include proper redirect URIs
- Fix MIME type for app files by preserving directory structure in sanitize_file_path()
- Add with_event_type() to TaskProgressEvent for correct WebSocket event types
- broadcast_manifest_update() now sends 'manifest_update' type correctly
- update_item_status() broadcasts automatically for real-time file progress
Database Schema v7.0.0:
- Create new 'gb' schema with PostgreSQL ENUMs instead of VARCHAR for all domain values
- Add sharding infrastructure (shard_config, tenant_shard_map tables)
- Implement partitioned tables for sessions, messages, and analytics (monthly partitions)
- Add Snowflake-like ID generation for distributed systems
- Design for billion-user scale with proper indexing strategies
Rust Enums:
- Add comprehensive enum types in core/shared/enums.rs
- Implement ToSql/FromSql for Diesel ORM integration
- Include: ChannelType, MessageRole, MessageType, LlmProvider, ContextProvider
- Include: TaskStatus, TaskPriority, ExecutionMode, RiskLevel, ApprovalStatus, IntentType
- All enums stored as SMALLINT for efficiency
Other fixes:
- Fix hardcoded gpt-4 model in auto_task modules to use bot config
- Add vector_db to required bootstrap components
- Add Qdrant health check before KB indexing
- Change verbose START messages to trace level
- Fix episodic memory role handling in Claude client
- Disable auth for /api routes during development
This is a DESTRUCTIVE migration - only for fresh installations.
- Fix match arms with identical bodies by consolidating patterns
- Fix case-insensitive file extension comparisons using eq_ignore_ascii_case
- Fix unnecessary Debug formatting in log/format macros
- Fix clone_from usage instead of clone assignment
- Fix let...else patterns where appropriate
- Fix format! append to String using write! macro
- Fix unwrap_or with function calls to use unwrap_or_else
- Add missing fields to manual Debug implementations
- Fix duplicate code in if blocks
- Add type aliases for complex types
- Rename struct fields to avoid common prefixes
- Various other clippy warning fixes
Note: Some 'unused async' warnings remain for functions that are
called with .await but don't contain await internally - these are
kept async for API compatibility.
- Add BOTSERVER_INSTALLERS_PATH env var to use local installers without downloading
- Replace hardcoded ./botserver-stack paths with configurable stack_path
- Add stack_dir() and vault_bin() helper methods in BootstrapManager
- Add Port: 8300 to Zitadel config to fix port binding issue
- Start Directory service before setup_directory() call
- Add SKIP_LLM_SERVER env var to skip local LLM in tests
- Update template loading to check ../bottemplates and botserver-templates paths
Instead of downloading pre-built binaries (which may require AVX2),
compile llama.cpp from source during installation. This ensures:
- Works on older CPUs (Sandy Bridge, Haswell, etc.)
- Uses GGML_NATIVE=ON to optimize for the current CPU
- Binary path updated to build/bin/llama-server
Reverts the AVX2 detection that was incorrectly disabling LLM.
- Add cpu_supports_avx2() function to check /proc/cpuinfo for AVX2 flag
- Skip LLM binary download on CPUs without AVX2 (pre-built llama.cpp requires it)
- Add runtime check for llama-server binary compatibility (catches 'Illegal instruction')
- Provide helpful error messages about alternatives (compile from source or use external API)
- Sandy Bridge (2nd gen i7) and older CPUs now gracefully skip local LLM
When llm-server=false in bot_configuration, the code now skips
attempting to start local llama-server processes. This prevents
the 60-attempt timeout error when using external LLM endpoints
or when local LLM serving is intentionally disabled.
- Add email-read-pixel config parameter to enable/disable tracking
- Implement tracking pixel injection in HTML emails
- Add sent_email_tracking table with migration
- Create 4 new API endpoints:
- GET /api/email/tracking/pixel/{id} - serve pixel & record read
- GET /api/email/tracking/status/{id} - get email read status
- GET /api/email/tracking/list - list all tracked emails
- GET /api/email/tracking/stats - get aggregate statistics
- Store tracking data: read_at, read_count, IP, user_agent
- Integrate with send_email() to auto-inject pixel when enabled
This commit introduces comprehensive documentation and implementation
for multi-agent orchestration capabilities:
- Add IMPLEMENTATION-PLAN.md with 4-phase roadmap
- Add Kubernetes deployment manifests (deployment.yaml, hpa.yaml)
- Add database migrations for multi-agent tables (6.1.1, 6.1.2)
- Implement A2A protocol for agent-to-agent communication
- Implement user memory keywords for cross-session persistence
- Implement model routing for dynamic L
The commit adds a complete example environment configuration file
documenting all available settings for BotServer, including logging,
database, server, drive, LLM, Redis, email, and feature flags.
Also removes hardcoded environment variable usage throughout the
codebase, replacing them with configuration via config.csv or
appropriate defaults. This includes:
- WhatsApp, Teams, Instagram adapter configurations
- Weather API key handling
- Email and directory service configurations
- Console feature conditionally compiles monitoring code
- Improved logging configuration with library suppression
The sqlx database library has been removed from the project along with
associated database-specific code that was no longer being used. This
includes removal of various sqlx-related dependencies from Cargo.lock
and cleanup of database connection pool references.
