Claudia MCP server
Task management for AI agents with hierarchical tasks, dependencies, sprints, and coordination.
15 stars43 downloads/wk
Reviews
Write oneNobody has reviewed Claudia yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
Claudia tools (48, 12 write)
write = sends, deletes, buys or postsRead from the package source without running it. The installed server may list more.
display_kanbanDisplay tasks in a Kanban board format with columns for each status.
display_legendDisplay the legend for status icons, priority indicators, and task types.
display_progressDisplay project progress summary with charts and statistics.
display_sprintDisplay a sprint as a detailed card with task breakdown and progress.
display_sprintsDisplay all sprints in a formatted list with progress indicators.
display_taskDisplay a single task as a detailed card with all information.
display_tasksDisplay tasks in a formatted CLI table view. Great for visual overview of task status.
project_createwrite actionCreate a new project. Projects group tasks and sprints together.
project_currentGet the currently selected project. If no project is selected, returns null with a list of available projects.
project_deletewrite actionDelete a project. Tasks and sprints in the project are orphaned (project_id set to null).
project_listList all projects.
project_readGet project details by ID.
project_selectSelect a project as the current working project. Subsequent task/sprint operations will be scoped to this project.
project_updatewrite actionUpdate project properties.
sprint_activateSet a sprint as the active sprint.
sprint_createwrite actionCreate a new sprint with optional date range.
sprint_deletewrite actionDelete a sprint (tasks are unassigned, not deleted).
sprint_listList all sprints with status and task counts.
sprint_showGet sprint details with associated tasks.
sprint_updatewrite actionUpdate sprint properties.
task_abandonAbandon a task and return it to pending status. Use when you cannot complete a task and need to release it. A reason is required and will be recorded in the task context. The task becomes available for other agents to claim.
task_blockedList tasks with unsatisfied dependencies.
task_claimCRITICAL: Use this when starting work on a task. Atomically claims the task AND automatically transitions it to 'in_progress'. You MUST call this before working on any task. Returns success status and the updated task.
task_context_getRetrieve stored agent context from a task.
task_context_mergeMerge context into existing task context (deep merge). Max 64KB total.
task_context_setStore agent context/memory on a task (overwrites existing). Max 64KB.
task_createwrite actionCreate a new task with optional parent, priority, acceptance criteria, and metadata. Returns the created task.
task_create_manywrite actionCreate multiple tasks atomically (max 100). Returns array of created tasks.
task_deletewrite actionDelete a task by ID. Children are orphaned (parent_id set to null).
task_dependenciesGet dependencies and dependents for a task.
task_dependency_addAdd a dependency between tasks (task depends on another).
task_dependency_removewrite actionRemove a dependency between tasks.
task_finishRECOMMENDED: Complete a task you're working on. This is the preferred way to mark a task as done. Optionally provide a completion summary. Use this instead of: task_transition(id, "completed") The task must be in_progress and claimed by you. Returns the completed task.
task_handoffHand off a task to another agent. Transfers ownership of the task while preserving its current status. Optionally include notes for the receiving agent. Handoff history is recorded in the task context.
task_listQuery tasks with filtering, sorting, and pagination. Returns array of tasks.
task_readRetrieve a single task by ID with full details including context and metadata.
task_readyList tasks ready to work on (all dependencies satisfied).
task_releaseRelease a claimed task. Only succeeds if task is claimed by the specified agent.
task_startRECOMMENDED: Create a new task AND immediately start working on it. This is the preferred way to create tasks you'll work on right away. Automatically sets status to "in_progress" and claims the task for you. Use this instead of: task_create + task_claim Returns the created task (already in_progress).
task_transitionChange task status with workflow validation. IMPORTANT: Use this to mark task as 'completed' when you finish working on it. Returns the updated task and transition details.
task_transition_manyTransition multiple tasks to a new status with validation.
task_treeGet hierarchical view of task and all descendants.
task_updatewrite actionUpdate one or more fields on an existing task. Returns the updated task.
task_update_manywrite actionUpdate multiple tasks with the same changes.
task_verification_statusGet verification status for a task's acceptance criteria. Returns all criteria with their verification status and overall progress.
task_verifyVERIFY: Mark an acceptance criterion as verified with optional evidence. Use this to record that you've verified a specific acceptance criterion on a task. Returns the updated criterion and overall verification progress. Criterion IDs are like 'ac-1', 'ac-2', etc. Use task_verification_status to see all criteria.
task_workspaceSESSION CONTEXT: Call this when starting a new session to understand your workspace. Returns: - Your session agent ID - Tasks currently in_progress (yours and orphaned) - Pending tasks ready to work on - Suggested next actions - Current project context This helps you resume work or pick up where another session left off.
workflow_infoWORKFLOW REFERENCE: PREFERRED: Use compound operations (task_start, task_finish) - they handle workflow automatically. ALTERNATIVE: If using separate operations: 1. STARTING WORK: Call task_claim first → auto-moves to "in_progress" 2. FINISHING WORK: Call task_transition(id, "completed") Note: Any mutation to a pending task now auto-claims it (implicit workflow). Returns the current workflow rules
Public scan report
scanner v0.1.10 · 2026-10-01 · same rubric, same numbers if you re-run it
- Code scan27 source files scanned13/25
- –Live reliabilityno gateway calls yet and no remote to proben/a
- –Tool poisoningtools not inspected (local package is not executed); not countedn/a
- Auth qualitylocal package, no credentials required12/15
- Maintenancelast push 215 days ago4/15
- Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
Findings (1)
- highShell command built from a string (injection risk)
exec.shell-concatsrc/cli/web-server.ts: …(process.platform === "darwin") { execSync(`open "${url}"`); } else if (process.platform…
How Claudia compares
Against every graded Productivity MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 57/100
- higher than 10% of 1,299 Productivity servers
- Tools
- 48
- the typical Productivity server has 7
- Tools that change things
- 25%
- typical: 0%. Fewer write tools means less an agent can break
- GitHub stars
- 15
- more than 81% of 854 Productivity servers
- Downloads a week
- 43
- more than 31% of 427 Productivity servers with a package
What its tools let an agent do: 36 read · 12 change
Install Claudia in Claude Code, Cursor or VS Code
Runs npx -y claudia-cli on your machine. Read the scan report first; the gateway never runs local packages.
claude mcp add claudia -- npx -y claudia-cli
What the publisher says
From the Claudia repository's README, as published. We do not edit it. Read it on GitHub
Claudia
A task management system built for AI agents.
Official website: https://claudiacli.com/
Claudia provides structured task tracking with a Model Context Protocol (MCP) server, enabling AI assistants like Claude to manage their own work through hierarchical tasks, dependencies, sprints, and acceptance criteria verification.
Why Claudia?
AI agents need a way to:
- Track progress across complex, multi-step tasks
- Coordinate when multiple agents work on the same project
- Remember context between sessions (64KB JSON storage per task)
- Verify work against acceptance criteria before completion
- Organize work into sprints and project hierarchies
Claudia provides all of this through both a CLI and MCP server interface.
Quick Start
# Install Bun if you haven't already
curl -fsSL https://bun.sh/install | bash
# Clone and build
git clone https://github.com/yuvalsuede/claudia.git
cd claudia
bun install
bun run build
# Initialize and start using
./claudia db init
./claudia task create --title "My first task"
./claudia task listTry the Demo
See Claudia in action with sample data:
# Seed demo project with sample tasks and sprints
bun run seed:demo
# Open the web dashboard
./claudia @@ --port 3333Then open http://localhost:3333 in your browser to explore the kanban board and sprint views.
Screenshots
Task Board
Sprint Management
Features
Installation
Prerequisites
- Bun runtime (v1.0+)
From Source
git clone https://github.com/yuvalsuede/claudia.git
cd claudia
bun installBuild Standalone Binary
bun run build
# Creates ./claudia binary
# Optional: install globally
cp claudia ~/.bun/bin/Usage
CLI Commands
Task Management
# Create a task
claudia task create --title "Implement feature X" --priority p1
# Create with acceptance criteria
claudia task create --title "Add login" --acceptance-criteria "Has email field" --acceptance-criteria "Has password field"
# List tasks
claudia task list
claudia task list --status in_progress --priority p0,p1
# Show task details
claudia task show <task-id>
# Update a task
claudia task update <task-id> --title "New title" --priority p0
# Transition status
claudia task transition <task-id> --to in_progress
# Delete a task
claudia task delete <task-id> --forceTask Hierarchy
# Create a subtask
claudia task create --title "Subtask" --parent <parent-id>
# View task tree
claudia task tree # Full tree
claudia task tree <task-id> # Subtree from taskTask Context (Agent Memory)
Shortened. The full README is on GitHub.
Nothing above is checked by us. What we check is on the safety report.
Claudia: common questions
- Is Claudia MCP server safe?
- With care: it is graded C, so read the findings first (57/100). Read the Claudia safety report
- How do I install Claudia?
- It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
- Does Claudia need an API key?
- Not as far as the registry entry and our scan can tell: no credentials are declared or required.
- Is Claudia maintained?
- The last commit was 249 days ago (2026-02-28). The latest release is v0.1.3.
- What can I use instead of Claudia?
- Servers from other publishers that do the same job: Agent Tasks MCP server.
- How does Claudia compare with other Productivity MCP servers?
- safety score 57/100 (higher than 10% of 1,299 Productivity servers); tools 48 (the typical Productivity server has 7); tools that change things 25% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.
Alternatives to Claudia
Same job from other publishers: the closest match first, then the best rated.
- Agent TasksPipeline task management for AI agents - stages, dependencies, artifacts, claimingnot reviewedGrowingA