How do I use Agentic Backlog MCP Server?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Agentic Backlog MCP Server show my open tasks" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

Agentic Backlog MCP Server

by hakenshi

Overview Schema Source Related Servers Discussions Admin

JavaScript

Local

Agentic Backlog MCP Server

Local-first MCP server for AI backlog management.

This package runs over stdio (Node.js) and forwards MCP tool calls to a running backlog API.

Exposed tools

backlog.identify_project
backlog.health
backlog.version
backlog.list_projects
backlog.get_project
backlog.get_kanban_url
backlog.create_task
backlog.list_tasks
backlog.get_task
backlog.find_tasks_by_title
backlog.update_task
backlog.update_task_by_title
backlog.delete_task
backlog.update_task_status
backlog.add_task_note
backlog.plan_from_context
backlog.get_focus
backlog.claim_task
backlog.release_task
backlog.restore_task
backlog.get_board
backlog.get_console_table

Related MCP server: boardagent-mcp

Requirements

Node.js 18+

Environment

BACKLOG_API_BASE_URL=http://127.0.0.1:38117/api
BACKLOG_REQUEST_TIMEOUT_MS=1800
BACKLOG_API_FAIL_FAST_MS=15000
BACKLOG_API_FAILURE_THRESHOLD=1

Run locally

npm install
npm run build
npm start

For development:

npm run dev

MCP config example

.mcp.json file:

{
  "mcp": {
     "agentic-backlog": {
      "command": "npx",
      "args": ["-y", "@hakenshi/agentic-backlog-mcp-server"],
      "env": {
        "BACKLOG_API_BASE_URL": "http://127.0.0.1:38117/api"
      }
    }
  }
}

Notes

This server uses stdio transport only.
Do not use console.log in MCP stdio mode (stdout breaks JSON-RPC). Logs must go to stderr.
backlog.delete_task requires explicit confirm: "DELETE".
backlog.plan_from_context is preview-only by default. Set apply: true to persist changes.
API resilience is fail-fast: when repeated timeout/5xx errors happen, the server opens a short circuit window and returns immediate 503 errors so agent runs do not stall.

Available Tools

22 tools

backlog.add_task_noteAdd task noteC

Adds a task note through the running backlog API.

Inputs: note (required), task_id (required), source, agent_id, +3 more.

backlog.claim_taskClaim taskB

Claims a task with TTL for an agent/session to avoid multi-agent collisions.

Inputs: task_id (required), agent_id (required), session_id (required), note, +3 more.

backlog.create_taskCreate taskC

Creates a task through the running backlog API.

Inputs: title (required), project_id (required), source, status, +6 more.

backlog.delete_taskDelete taskC

Deletes a task through the running backlog API.

Inputs: confirm (required), task_id (required), idempotency_key.

backlog.find_tasks_by_titleFind tasks by titleC

Finds tasks in a project by title keywords using backlog API task list.

Inputs: query (required), project_id (required), limit.

backlog.get_boardGet board snapshotC

Returns board grouped snapshot from the running backlog API.

Inputs: project_id (required).

backlog.get_console_tableGet board as console tableC

Returns a console-table style board snapshot from backlog API.

Inputs: project_id (required), limit.

backlog.get_focusGet focusB

Returns what matters now for a project: top priority, blocked, stale, and in progress tasks.

Inputs: project_id (required), stale_hours.

backlog.get_kanban_urlGet kanban URLC

Returns a browser URL for visual kanban from the running backlog API.

Inputs: project_id (required), base_url.

backlog.get_projectGet projectB

Returns project metadata and board summary from the running backlog API.

Inputs: project_id (required).

backlog.get_taskGet taskA

Returns a single task by id from the running backlog API.

Inputs: task_id (required).

backlog.healthBacklog healthA

Checks backlog API health and connectivity before mutations.

Inputs: no named parameters.

backlog.identify_projectIdentify or create projectB

Detects git context and resolves/creates a project in the running backlog API (Docker app source of truth).

Inputs: cwd, name, description.

backlog.list_projectsList projectsC

Lists projects from the running backlog API.

Inputs: limit.

backlog.list_tasksList tasksC

Lists project tasks from the running backlog API.

Inputs: project_id (required), limit, status.

backlog.plan_from_contextPlan tasks from contextC

Delegates planning to running backlog API planner endpoint.

Inputs: context (required), project_id (required), apply, source, +4 more.

backlog.release_taskRelease task claimC

Releases an active task claim for the given agent/session.

Inputs: task_id (required), agent_id (required), session_id (required), source, +1 more.

backlog.restore_taskRestore taskC

Restores a soft-deleted task within restore window.

Inputs: task_id (required), source, idempotency_key.

backlog.update_taskUpdate taskC

Updates task fields through the running backlog API.

Inputs: task_id (required), title, reason, source, +8 more.

backlog.update_task_by_titleUpdate task by titleC

Finds a task by title query and updates it through the backlog API.

Inputs: query (required), project_id (required), title, reason, +9 more.

backlog.update_task_statusUpdate task statusC

Moves a task between states through the running backlog API.

Inputs: status (required), task_id (required), reason, source, +4 more.

backlog.versionBacklog versionA

Returns API version and protocol compatibility information.

Inputs: no named parameters.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

Deploy Server

Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

TDQS

C2.9/5.0

Scored 2026-09-20 14:25 across 22 tools

Disambiguation3/5

Many tools are distinct, but there are several overlapping boundaries: get_board and get_console_table both return board snapshots, update_task/update_task_by_title/update_task_status all mutate tasks, and get_focus/list_tasks overlap as list-like views. Descriptions help, but an agent could easily pick the wrong one for state changes or snapshot retrieval.

Naming Consistency4/5

Names consistently use the backlog. prefix with snake_case verb_noun construction (list_tasks, create_task, update_task). Minor deviations are bare nouns health and version and the looser get_/list_ distinction, but the overall pattern is predictable and readable.

Tool Count3/5

22 tools is on the heavy side and sits in the 16-25 borderline range. Many could be consolidated (board formats, update variants, health/version), though the broad project/task/planning/claim domain explains some of the volume.

Completeness4/5

The surface covers task CRUD well, including restore, status transitions, notes, claims, board snapshots, focus, and planning. Minor gaps exist such as no project update/delete and limited search beyond title keywords, but core agent workflows do not hit dead ends.

Maintenance

ActivityInactive

ResponsivenessNo recent community issues

Resources

Related MCP Connectors

AI Tasker
dev.theboomer
Task and sprint coordination for humans and AI agents, with scoped API keys and full audit trail.
Project Management Security Autonomous Agents
1
https://mcp.ai-tasker.theboomer.dev/mcp
AI Agent Background Job Queue
io.github.Ansarii
Durable background job execution, async task scheduling, and state persistence for AI agents.
Autonomous Agents Agent Orchestration
1
https://neon_innovation_lab--mcp-durable-tasks-job-queue-backend.apify.actor/mcp
Masaro
ai.masaro
Task management for teams building with AI agents. Agents claim tasks and report progress.
Project Management Agent Orchestration
2
https://masaro.ai/api/mcp/mcp
LTD MCP
io.github.Eshller
Project management MCP for AI agents with safe task reads and writes.
Project Management Agent Orchestration
12
https://api.eshway.com/mcp/

Related MCP Servers

sprinter
Project Management Autonomous Agents
danimoya
A
license
Not graded
quality
C
maintenance
A self-hosted backlog tracker with priority scoring and an MCP server, enabling AI agents to autonomously pull, work on, and update tasks via JSON-RPC tools.
Updated 2 months ago (2026-09-10 11:57 UTC)
MIT
boardagent-mcp
Project Management Workplace & Productivity
0xAngel-Sec
A
license
A
quality
B
maintenance
Enables AI agents to manage tasks on a local-first board via MCP, exposing task creation, updates, and queries through a thin adapter over the REST service.
Updated 2 months ago (2026-08-28 10:14 UTC)
7
MIT
todo-mcp
Project Management Workplace & Productivity
Karsten-Larson
F
license
Not graded
quality
C
maintenance
Enables AI assistants to manage a local-first Kanban board with Azure DevOps-style workflow rules, including task creation, status updates, subtask breakdown, full-text search, daily summaries, and next-task suggestions.
Updated 2 months ago (2026-09-07 18:32 UTC)
-
project-tasks-mcp
Project Management Developer Tools Autonomous Agents
augustoOliveira1993
F
license
Not graded
quality
B
maintenance
Enables managing projects, features, tasks, dependencies, executions, and human review through a Streamable HTTP endpoint, along with persistent inter-agent cooperation via task messaging and event subscriptions. It also supports administrative actions such as approvals, cancellations, unblocking, and member management.
Updated 14 days ago (2026-10-27 12:05 UTC)
-

Agentic Backlog MCP Server

Exposed tools

Requirements

Environment

Run locally

MCP config example

Notes

Available Tools

Tool Schema Changelog

TDQS

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

sprinter

boardagent-mcp

todo-mcp

project-tasks-mcp