Conductor TasksCAUTION
A task management system designed for AI development
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://opensource.org/licenses/MIT) [](https://badge.fury.io/js/conductor-tasks)
Transform requirements into actionable tasks, generate implementation plans, track progress, and accelerate development – all powered by AI, directly within your workflow.
Conductor Tasks is an intelligent assistant designed for developers. It integrates seamlessly into your editor (via MCP) or works as a standalone CLI tool, leveraging multiple LLMs to streamline your development process from planning to execution.
Key Features
- AI-Powered Task Generation: Instantly parse Product Requirements Documents (PRDs), markdown files, or even unstructured notes into structured, actionable tasks.
- Intelligent Task Expansion & Planning: Automatically break down complex tasks into detailed subtasks and generate step-by-step implementation plans using context-aware AI.
- Powerful CLI for Automation: Leverage a comprehensive command-line interface for scripting, automation, and use outside of an editor.
- Versatile Task Templating: Create new tasks from predefined or custom templates, standardizing common workflows and saving setup time.
- Visual Task Management: Get a clear overview of your project with Kanban boards, dependency trees, and summary dashboards
- Multi-Provider LLM Flexibility: Works out-of-the-box with OpenAI, Anthropic, Groq, Mistral, Google Gemini, Perplexity, xAI, Azure OpenAI. Easily configure custom/local OpenAI-compatible endpoints (like Ollama or LM Studio). You're not locked into a single provider – choose the best LLM for each specific need.
Why Conductor Tasks?
While many AI-powered task management tools offer valuable assistance, Conductor Tasks is engineered to pro
96df0096e974OBSERVED · 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. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add conductor-tasks --env ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY} --env CLAUDE_API_KEY=${CLAUDE_API_KEY} --env GEMINI_API_KEY=${GEMINI_API_KEY} --env GROQ_API_KEY=${GROQ_API_KEY} -- npx -y [email protected]{
"mcpServers": {
"conductor-tasks": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"ANTHROPIC_API_KEY": "${ANTHROPIC_API_KEY}",
"CLAUDE_API_KEY": "${CLAUDE_API_KEY}",
"GEMINI_API_KEY": "${GEMINI_API_KEY}",
"GROQ_API_KEY": "${GROQ_API_KEY}"
}
}
}
}Exposed tools (26)
20 read · 5 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
.roomode | read | Roo Code Mode Configuration |
.windsurfrules | read | Windsurf IDE Rules |
add-task-note | write | Add a note, progress update, or comment to a task |
conductor-tasks.rules | read | Cline IDE Rule: conductor-tasks.rules |
create-task | write | Create a new task with details |
create-task-from-template | write | Create a new task from a template |
delete-task | destructive | Delete a task |
expand-task | read | Expand a task with more detailed information and subtasks |
generate-diff | read | Generate a diff for a file based on a change description. |
generate-implementation-steps | read | Generate detailed implementation steps for a task |
get-next-task | read | Get the next task to work on |
get-task | read | Get details of a specific task |
get-task-template | read | Get details of a specific task template |
help-implement-task | read | Get AI assistance to implement a specific task |
initialize-project | read | Initialize the project, including task management (TASKS.md) and IDE rules. |
list-task-templates | read | List available task templates |
list-tasks | read | Get a list of tasks with filtering and sorting options |
parse-prd | write | Parse a PRD (Product Requirements Document) and create tasks from it |
parse-prd-file | read | Parse a PRD file from disk and extract tasks |
propose-diff | read | Propose a diff to be applied to a file. Currently acknowledges only; does not apply. |
research-topic | read | Research a topic using available LLM capabilities (e.g., Perplexity, tool-calling for search). |
suggest-task-improvements | read | Get AI suggestions for improving a task |
update-task | write | Update an existing task |
visualize-tasks-dashboard | read | Display task dashboard with summary statistics |
visualize-tasks-dependency-tree | read | Display task dependency tree |
visualize-tasks-kanban | read | Display tasks in a Kanban board view |
Trust audit
CAUTIONgrade B · trust 89/100 Install with care. The audit found things worth knowing before you trust its output.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | WARN |
| 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
- declared (5 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (9)
│ └── [!!] #126 Create user profile page 👨💻
Status: ✅ Done 🕒 Todo 👨💻 In Progress 👀 Review ⛔ Blocked
delete-task
import { ErrorHandler, ErrorCategory, ErrorSeverity, TaskError } from '../../core/errorHandler.js';import { JsonUtils } from '../../core/jsonUtils.js';import { LLMProvider, LLMProviderConfig, LLMRequest, LLMResponse } from '../../core/types.js';import { JsonUtils } from '../../core/jsonUtils.js';import { ErrorHandler, ErrorCategory, ErrorSeverity, TaskError } from '../../core/errorHandler.js';@anthropic-ai/sdk, @google/generative-ai, @mistralai/mistralai, @modelcontextprotocol/sdk, ajv, chalk, commander, diff
Gates applied: no_behavioural_pass.
96df0096e974full audit observations/trust-audit/mcp-server/hridaya423__conductor-tasks.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 | 96df0096e974 | CAUTION | B | 89 | first audit |
Questions
What is the Conductor Tasks MCP server?
A task management system designed for AI development
What tools does Conductor Tasks expose?
26 in total: 20 read-only, 5 that write, and 1 that can delete or overwrite (delete-task). Every one is listed on this page with its risk.
Is Conductor Tasks safe to connect to an agent?
With care. The audit graded it B (89/100) and found 9 things worth knowing before you trust this server, listed below with the exact line each was found on. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Conductor Tasks need?
It reads ANTHROPIC_API_KEY, CLAUDE_API_KEY, GEMINI_API_KEY, GROQ_API_KEY, MAX_TOKENS, MISTRAL_API_KEY, MIXTRAL_API_KEY, OLLAMA_API_KEY, OPENAI_API_KEY, OPENROUTER_API_KEY, PERPLEXITY_API_KEY and XAI_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does Conductor Tasks run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as conductor-tasks at 0.2.4.
How current is this page?
The grade is for one exact copy of the source (96df0096e974), read on 2026-10-07. The repository is watched and re-audited when it changes.