This project is a plugin for LM Studio that provides a rich set of tools to a large language model. It acts as a bridge between the LLM and your local environment, enabling autonomous coding, research, and file management.
[!IMPORTANT]
LM Studio does NOT support automatic updates. If you encounter issues, please try manually updating first by removing the current version and re-downloading from the plugin website. LM Studio may show an "already installed" tooltip, even if your version is outdated.
Key Features
File System Mastery
Full Control: Create, read, update, delete, move, and copy files.
Safe & Secure: All operations are sandboxed to your workspace directory to prevent path traversal attacks.
Smart Updates: Use replace_text_in_file to make surgical edits instead of rewriting large files.
๐ ๏ธ Tool Reordering & Optimization: Reordered the tools list to prioritize casual utilities, improving agent tool selection. Upgraded the web search tool with smart Chrome detection and fallback.
๐ค Sub-Agent Reliability Rework: Overhauled the sub-agent loop to prevent infinite loops, improved tool call parsing, unified path/content normalization, and added explicit task completion/abortion capabilities (TASK_FAILED).
โจ New Sub-Agent Tools: Empowered sub-agents with multi_replace_text, , and background command execution. Main agents now support robust batch file saving.
Older Updates (v1.3.1 & earlier)
v1.3.1
๐ Full Internationalization (i18n): Added complete UI and runtime support for English, German, , and .
Autonomous Agents
Secondary Agent: Delegate complex tasks (coding, summarization) to a second local model/server with support for the main model already used and loaded by LM Studio!
Code Execution
Sandboxed: Run JavaScript (Deno) and Python code.
Terminal: Execute shell commands or open real terminal windows for interactive tasks.
[!WARNING]
Enabling shell or terminal execution allows the model to run arbitrary commands on your system. If enabled, the model may be able to modify files and escape the sandbox environment.
Web & RAG
Research: Search DuckDuckGo, Wikipedia, or fetch raw web content.
Advanced Browser Navigation: Persistent browser_session_open/control/close flow for multi-step browsing and automation.
Web RAG: Chat with website content.
Local RAG: Semantic search over your workspace files (rag_local_files).
๐ก Tip: Need persistent long-term memory for your agent?
Check out my other project: Local Memory MCP โ A privacy-first memory server with knowledge graph support.
Installation
The plugin can be installed using the following link:
This project is a plugin for LM Studio that provides a rich set of tools to a large language model. It acts as a bridge between the LLM and your local environment, enabling autonomous coding, research, and file management.
[!IMPORTANT]
LM Studio does NOT support automatic updates. If you encounter issues, please try manually updating first by removing the current version and re-downloading from the plugin website. LM Studio may show an "already installed" tooltip, even if your version is outdated.
Key Features
File System Mastery
Full Control: Create, read, update, delete, move, and copy files.
Safe & Secure: All operations are sandboxed to your workspace directory to prevent path traversal attacks.
Smart Updates: Use replace_text_in_file to make surgical edits instead of rewriting large files.
๐ ๏ธ Tool Reordering & Optimization: Reordered the tools list to prioritize casual utilities, improving agent tool selection. Upgraded the web search tool with smart Chrome detection and fallback.
๐ค Sub-Agent Reliability Rework: Overhauled the sub-agent loop to prevent infinite loops, improved tool call parsing, unified path/content normalization, and added explicit task completion/abortion capabilities (TASK_FAILED).
โจ New Sub-Agent Tools: Empowered sub-agents with multi_replace_text, , and background command execution. Main agents now support robust batch file saving.
Older Updates (v1.3.1 & earlier)
v1.3.1
๐ Full Internationalization (i18n): Added complete UI and runtime support for English, German, , and .
Autonomous Agents
Secondary Agent: Delegate complex tasks (coding, summarization) to a second local model/server with support for the main model already used and loaded by LM Studio!
Code Execution
Sandboxed: Run JavaScript (Deno) and Python code.
Terminal: Execute shell commands or open real terminal windows for interactive tasks.
[!WARNING]
Enabling shell or terminal execution allows the model to run arbitrary commands on your system. If enabled, the model may be able to modify files and escape the sandbox environment.
Web & RAG
Research: Search DuckDuckGo, Wikipedia, or fetch raw web content.
Advanced Browser Navigation: Persistent browser_session_open/control/close flow for multi-step browsing and automation.
Web RAG: Chat with website content.
Local RAG: Semantic search over your workspace files (rag_local_files).
๐ก Tip: Need persistent long-term memory for your agent?
Check out my other project: Local Memory MCP โ A privacy-first memory server with knowledge graph support.
Installation
The plugin can be installed using the following link:
Batch Processing:save_file supports creating multiple files in one go.
Cleanup: Use delete_files_by_pattern to wipe temporary files instantly.
search_directory
Simplified Chinese
Traditional Chinese
๐ Dual-Layer Translation: Supports both "Config UI" (static) and "Agent Messages" (dynamic runtime) languages.
๐ UI Language Override: Added a field to manually force the Config UI into a specific locale for testing, bypassing OS-level detection on the next plugin restart.
v1.3.0
๐ Native GitHub CLI Tools: Added gh_auth, gh_create_issue, gh_list_issues, gh_view_comments, gh_create_pr, gh_list_prs, gh_view_pr_diff, and gh_push for secure, structured GitHub interactions without generic shell commands
๐ฟ Enhanced Git Workflow: Added git_add and git_checkout tools to complete the native git toolchain (status, diff, log, commit, add, checkout)
โ๏ธ New Sidebar Toggle: Added "Allow GitHub CLI Tools" setting to enable/disable gh_* tools independently
๐ก๏ธ Dependency Guards: All CLI tools now verify installation before execution and provide clear prompts if missing
v1.2.0
๐ก๏ธ Sub-Agent Tool Validation: Added early parameter validation with clear error messages to prevent silent failures when sub-agents use wrong parameter names or absolute paths outside workspace
๐งช Regression Tests: Added 14 new tests for tool call validation logic (total: 51 tests)
๐ฌ Better Error Feedback: Sub-agents now receive TOOL_VALIDATION_ERROR messages with helpful hints about correct parameter names vs common mistakes
v1.1.1 (2026-04-08)
Browser Reliability and Navigation Context
Fixed: Browser action clicks now include a DOM-level fallback when Puppeteer reports "Node is either not clickable or not an Element"
Improved: Browser click actions now retry native click after ~300ms before falling back
Added:browser_session_open returns full page text by default (include_page_text defaults to true)
Fixed: Gemma 4 and other models using {"tool": "...", "parameters": {...}} format now work correctly with consult_secondary_agent
Added: Advanced browser navigation (browser_session_open, browser_session_control, browser_session_close) including in-page fuzzy find and URL-change notices
Added: Structured sub-agent handoff message support (handoff_message) for relay/summary workflows
Added: Enable Sub-Agent Debug Logging toggle in plugin settings
Added: Support for direct {file_name, content} JSON format from some models
v1.0.x Previous Updates
Smart Context Injection:subagent_docs.md is automatically loaded into the context
Enhanced Reporting: Fixed file path reporting in consult_secondary_agent
Project Tracking: Sub-agents enforce creation of beledarian_info.md
Strict Naming: Improved instructions for correct file extensions
Auto-Save: When the sub-agent generates code, the system automatically detects and saves it to your disk. No more copy-pasting!
Auto-Debug: (Optional) Triggers a "Reviewer" agent to analyze generated code and fix bugs automatically before returning the result.
Structured Handoff: Sub-agents can return a dedicated handoff_message for the main agent to relay findings/research.
Project Context: Agents can read beledarian_info.md to understand your project's history.
npm install
Run in development mode:
From within the project directory, run the following command:
lms dev
This will start the plugin in development mode. LM Studio should automatically pick it up. Any changes you make to the source code will cause the plugin to automatically reload.
Enable Secondary Agent: Unlock the power of sub-agents.
Sub-Agent Profiles: Custom prompts for "Coder", "Reviewer", etc.
Show Full Code Output: Toggle whether to display the full code in chat or hide it for brevity (files are still saved).
Default Workspace Path: Set the startup workspace directory used by the plugin.
Safety: Enable/Disable "Allow Code Execution" for Python/JS/Shell.
Browser Safety: Browser automation for sub-agents requires all three toggles: Allow Browser Control + Sub-Agent: Allow Web Search + Sub-Agent: Allow Browser Control.
Batch Processing:save_file supports creating multiple files in one go.
Cleanup: Use delete_files_by_pattern to wipe temporary files instantly.
search_directory
Simplified Chinese
Traditional Chinese
๐ Dual-Layer Translation: Supports both "Config UI" (static) and "Agent Messages" (dynamic runtime) languages.
๐ UI Language Override: Added a field to manually force the Config UI into a specific locale for testing, bypassing OS-level detection on the next plugin restart.
v1.3.0
๐ Native GitHub CLI Tools: Added gh_auth, gh_create_issue, gh_list_issues, gh_view_comments, gh_create_pr, gh_list_prs, gh_view_pr_diff, and gh_push for secure, structured GitHub interactions without generic shell commands
๐ฟ Enhanced Git Workflow: Added git_add and git_checkout tools to complete the native git toolchain (status, diff, log, commit, add, checkout)
โ๏ธ New Sidebar Toggle: Added "Allow GitHub CLI Tools" setting to enable/disable gh_* tools independently
๐ก๏ธ Dependency Guards: All CLI tools now verify installation before execution and provide clear prompts if missing
v1.2.0
๐ก๏ธ Sub-Agent Tool Validation: Added early parameter validation with clear error messages to prevent silent failures when sub-agents use wrong parameter names or absolute paths outside workspace
๐งช Regression Tests: Added 14 new tests for tool call validation logic (total: 51 tests)
๐ฌ Better Error Feedback: Sub-agents now receive TOOL_VALIDATION_ERROR messages with helpful hints about correct parameter names vs common mistakes
v1.1.1 (2026-04-08)
Browser Reliability and Navigation Context
Fixed: Browser action clicks now include a DOM-level fallback when Puppeteer reports "Node is either not clickable or not an Element"
Improved: Browser click actions now retry native click after ~300ms before falling back
Added:browser_session_open returns full page text by default (include_page_text defaults to true)
Fixed: Gemma 4 and other models using {"tool": "...", "parameters": {...}} format now work correctly with consult_secondary_agent
Added: Advanced browser navigation (browser_session_open, browser_session_control, browser_session_close) including in-page fuzzy find and URL-change notices
Added: Structured sub-agent handoff message support (handoff_message) for relay/summary workflows
Added: Enable Sub-Agent Debug Logging toggle in plugin settings
Added: Support for direct {file_name, content} JSON format from some models
v1.0.x Previous Updates
Smart Context Injection:subagent_docs.md is automatically loaded into the context
Enhanced Reporting: Fixed file path reporting in consult_secondary_agent
Project Tracking: Sub-agents enforce creation of beledarian_info.md
Strict Naming: Improved instructions for correct file extensions
Auto-Save: When the sub-agent generates code, the system automatically detects and saves it to your disk. No more copy-pasting!
Auto-Debug: (Optional) Triggers a "Reviewer" agent to analyze generated code and fix bugs automatically before returning the result.
Structured Handoff: Sub-agents can return a dedicated handoff_message for the main agent to relay findings/research.
Project Context: Agents can read beledarian_info.md to understand your project's history.
npm install
Run in development mode:
From within the project directory, run the following command:
lms dev
This will start the plugin in development mode. LM Studio should automatically pick it up. Any changes you make to the source code will cause the plugin to automatically reload.
Enable Secondary Agent: Unlock the power of sub-agents.
Sub-Agent Profiles: Custom prompts for "Coder", "Reviewer", etc.
Show Full Code Output: Toggle whether to display the full code in chat or hide it for brevity (files are still saved).
Default Workspace Path: Set the startup workspace directory used by the plugin.
Safety: Enable/Disable "Allow Code Execution" for Python/JS/Shell.
Browser Safety: Browser automation for sub-agents requires all three toggles: Allow Browser Control + Sub-Agent: Allow Web Search + Sub-Agent: Allow Browser Control.
