Agentic ToolsSAFE
A comprehensive Model Context Protocol (MCP) server providing AI assistants with powerful task management and agent memories capabilities with project-specific storage.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://badge.fury.io/js/@pimzino%2Fagentic-tools-mcp) [](https://www.npmjs.com/package/@pimzino/agentic-tools-mcp) [](https://github.com/Pimzino/agentic-tools-mcp/stargazers) [](https://github.com/Pimzino/agentic-tools-mcp/blob/main/LICENSE) [](https://nodejs.org/)
A comprehensive Model Context Protocol (MCP) server providing AI assistants with powerful advanced task management and agent memories capabilities with project-specific storage.
🔗 Ecosystem
This MCP server is part of a complete task and memory management ecosystem:
- 🖥️ [VS Code Extension](https://github.com/Pimzino/agentic-tools-mcp-companion) - Beautiful GUI interface for managing tasks and memories directly in VS Code
- ⚡ MCP Server (this repository) - Advanced AI agent tools and API for intelligent task management
💡 Pro Tip: Use both together for the ultimate productivity experience! The VS Code extension provides a visual interface while the MCP server enables AI assistant integration with advanced features like PRD parsing, task recommendations, and research capabilities.
Features
🎯 Advanced Task Management System with Unlimited Hierarchy (v1.8.0)
- Projects: Organize work into distinct projects with descriptions
- Unified Task Model: Single task interface supporting unlimited nesting depth
- Unlimited Hierarchy: Tasks → Subtasks → Sub-subtasks → infinite depth nesting
- Rich Features at All Levels: Every task gets priority, complexity, dependencies, tags, and time tracking
- Parent-Child Relationships: Flexible hi
8d6eee77e69bOBSERVED · 2026-10-07Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.
claude mcp add agentic-tools-mcp -- npx -y @pimzino/[email protected]
{
"mcpServers": {
"agentic-tools-mcp": {
"command": "npx",
"args": [
"-y",
"@pimzino/[email protected]"
]
}
}
}Exposed tools (29)
16 read · 9 write · 4 destructive. Blast radius: 4 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
analyze_task_complexity | read | Analyze task complexity and suggest breaking down overly complex tasks into smaller, manageable subtasks. Intelligent complexity analysis feature for better productivity and progress tracking. |
create_memory | write | Create a new memory in the agent memories system |
create_project | write | Create a new project in the task management system |
create_subtask | write | Create a new subtask within a specific task |
create_task | write | Create a new task within a specific project. Supports unlimited nesting depth - set parentId to create subtasks, sub-subtasks, etc. Leave parentId empty for top-level tasks. |
delete_memory | destructive | Delete a memory by ID (requires confirmation) |
delete_project | destructive | Delete a project and all its associated tasks and subtasks. This action cannot be undone. |
delete_subtask | destructive | Delete a subtask. This action cannot be undone. |
delete_task | destructive | Delete a task and all its associated subtasks. This action cannot be undone. |
generate_research_queries | read | Generate intelligent, targeted web search queries for task research. Provides structured search strategies to help AI agents find the most relevant information efficiently. |
get_memory | read | Get a specific memory by its ID |
get_next_task_recommendation | read | Get intelligent recommendations for the next task to work on based on dependencies, priorities, complexity, and current project status. Smart task recommendation engine for optimal workflow management. |
get_project | read | Get detailed information about a specific project by its ID |
get_subtask | read | Get detailed information about a specific subtask by its ID |
get_task | read | Get detailed information about a specific task by its ID |
infer_task_progress | read | Analyze the codebase to infer which tasks appear to be completed based on code changes, file creation, and implementation evidence. Intelligent progress inference feature for automatic task completion tracking. |
list_memories | read | List memories with optional filtering by category and limit |
list_projects | read | View all projects in the task management system |
list_subtasks | read | View subtasks, optionally filtered by task ID or project ID |
list_tasks | read | View tasks in hierarchical tree format. Filter by projectId (required) and/or parentId. Use parentId=null for top-level tasks, or specific parentId for subtasks. |
migrate_subtasks | read | Migrate existing subtasks to the unified task model. This tool converts all subtasks to tasks with parentId for unlimited nesting depth. Run this once after upgrading to ensure data compatibility. |
move_task | write | Move a task to a different parent in the hierarchy. Set newParentId to move under another task, or leave empty to move to top level. Supports unlimited nesting depth. |
parse_prd | read | Parse a Product Requirements Document (PRD) and automatically generate structured tasks with dependencies, priorities, and complexity estimates. This tool analyzes PRD content and creates a comprehensive task breakdown with intelligent analysis. |
research_task | read | Guide the AI agent to perform comprehensive web research for a task, with intelligent research suggestions and automatic memory storage of findings. Combines web research capabilities with local knowledge caching. |
search_memories | read | Search memories by text content matching |
update_memory | write | Update an existing memory\ |
update_project | write | Update the name and/or description of an existing project |
update_subtask | write | Update the name, details, and/or completion status of an existing subtask |
update_task | write | Update task properties including name, details, parent relationship (parentId), completion status, dependencies, priority, complexity, status, tags, and time estimates. Use parentId to move tasks within the hierarchy. |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (6 observation(s))
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (7)
delete_memory, delete_project, delete_subtask, delete_task
import { MemoryStorage } from '../../storage/storage.js';import { Memory } from '../../models/memory.js';import { MemoryStorage } from '../../storage/storage.js';import { MemoryStorage } from '../../storage/storage.js';import { MemoryStorage } from '../../storage/storage.js';@modelcontextprotocol/sdk, zod, @types/node, typescript
Gates applied: no_behavioural_pass.
8d6eee77e69bfull audit observations/trust-audit/mcp-server/pimzino__agentic-tools.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-07 | 8d6eee77e69b | SAFE | B | 89 | first audit |
Questions
What is the Agentic Tools MCP server?
A comprehensive Model Context Protocol (MCP) server providing AI assistants with powerful task management and agent memories capabilities with project-specific storage.
What tools does Agentic Tools expose?
29 in total: 16 read-only, 9 that write, and 4 that can delete or overwrite (delete_memory, delete_project, delete_subtask, delete_task). Every one is listed on this page with its risk.
Is Agentic Tools safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B. Separately from the audit: 4 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Agentic Tools need?
No credential environment variables were found in its source, so it appears to need none.
How does Agentic Tools run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as @pimzino/agentic-tools-mcp at 1.8.1.
How current is this page?
The grade is for one exact copy of the source (8d6eee77e69b), read on 2026-10-07. The repository is watched and re-audited when it changes.