PinkRoosterMcp

git clone https://github.com/pinkroosterai/PinkRoosterMcp.git cd PinkRoosterMcp make install

That's it.make installpulls the pre-built image from Docker Hub, registers the MCP server in Claude Code, installs PM skills, and starts all containers. No accounts, no configuration, no build step.

Every other PM-related MCP server is a wrapper around an existing SaaS product. PinkRoosterMcp ispurpose-built for AI agents— the data model, response format, and workflow skills are designed for AI consumption first, with a dashboard for human visibility.

Traditional workflow: 1. Open browser → find Jira ticket → read requirements 2. Switch to IDE → write code → run tests 3. Switch to browser → update ticket status → add comment 4. Repeat 50x per day

You: /pm-next --auto Agent: picks highest priority work → scaffolds plan → implements code → runs tests → commits → updates all project state → picks next item → repeats until done You: ☕

Complete a task and everything upstream updates automatically — no manual ticket grooming:

Task completed └→ Phase auto-completes (all tasks done) └→ Work Package auto-completes (all phases done) ├→ Linked Issue auto-resolved └→ Linked Feature Request auto-completed

Dependencies auto-block and auto-unblock. The agent always knows what happened downstream via structuredOperationResultresponses with state change notifications.

A single command drives the full development lifecycle:

The agent picks the highest-priority work item, scaffolds a plan if needed, implements every task (reading code, writing changes, running tests), commits after each work package, and loops until all work is done.

Every tool is designed for AI consumption — compact responses, actionable next steps, and state cascade notifications. Scaffold an entire work package with phases, tasks, and dependencies in a single call.

High-level slash commands that encode sophisticated PM workflows:/pm-scaffoldanalyzes your codebase to produce realistic target files and implementation notes./pm-implementhandles dependency ordering, test running with auto-fix, and phase verification gates./pm-auditruns parallel analysis agents across quality, security, performance, and architecture domains./pm-cleanupfinds dead code and scaffolds fix tasks.

Persistent knowledge store for decisions, patterns, and context that survives across agent sessions. Merge-by-name semantics — writing to an existing memory appends content and unions tags.

proj-1-issue-3,proj-1-wp-2-task-5— readable by both agents and humans in conversation. No GUIDs, no opaque numeric IDs.

Multi-User Accounts & Per-Project RBAC

Built-in user account system with Argon2id password hashing, session management, and per-project role-based access control. Four roles (SuperUser, Admin, Editor, Viewer) give fine-grained control over who can do what on each project. First user auto-gets SuperUser. API key auth for MCP/programmatic access runs in parallel — no RBAC overhead for agents.

Two layers of audit: per-entity field-change audit logs (old/new values on every field) and HTTP request activity logging (method, path, status, duration, caller). The dashboard surfaces both.

A real-time React dashboard lets you see everything your AI agent is tracking — at a glance or in full detail.

The dashboard home shows active counts, completion percentages, and priority next actions across all entity types.

Filter and sort issues by severity, priority, type, and state. Summary cards show active/inactive/terminal breakdowns with mini donut charts.

Every entity has a structured detail page with inline editing, state management, related entities, timeline, and a collapsible audit log.

Work packages are the execution unit — each one contains phases and tasks with dependency tracking and automatic state propagation.

Track ideas from proposal through completion with an 8-state lifecycle. Link feature requests to work packages to connect "what" to "how."

Every API request is logged with method, path, status, duration, and caller identity — giving you full observability into agent behavior.

┌─────────────────┐ MCP (Streamable HTTP) ┌─────────────────┐ │ Claude Code / │ ◄──────────────────────────► │ MCP Server │ │ Cursor / etc. │ │ :5200 │ └─────────────────┘ └────────┬────────┘ │ HTTP ┌─────────────────┐ HTTP / REST ┌────────▼────────┐ │ Dashboard │ ◄──────────────────────────► │ REST API │ │ :3000 │ │ :5100 │ └─────────────────┘ └────────┬────────┘ │ EF Core ┌────────▼────────┐ │ PostgreSQL │ │ :5432 │ └─────────────────┘

- The MCP server calls the API over HTTP — no shared database access, clean separation
- Human-readable IDs everywhere (proj-1-issue-3,proj-1-wp-2-task-5) — easy for both agents and humans
- State cascades automatically — completing all tasks in a phase completes the phase, which can complete the work package, which can complete linked issues and feature requests
- Full audit trail — every field change is recorded with old/new values

PinkRoosterMcp exposes 24 MCP tools organized by entity:

All write operations return structuredOperationResultJSON with state change cascades, so the agent always knows what happened downstream.

Sixteen Claude Code slash commands provide high-level project management workflows on top of the MCP tools:

Skills automatically propagate state to related entities. Starting a task activates its work package and linked issue/FR. Completing all tasks cascades completion upward through phases, work packages, and linked entities.

git clone https://github.com/pinkroosterai/PinkRoosterMcp.git cd PinkRoosterMcp make install

- Create.envwith working defaults (no edits needed)
- Register the MCP server in Claude Code (global scope, skipped if CLI not found)
- Install PM workflow skills to~/.claude/skills/
- Pull and start the pre-built Docker Hub image (no local build required)

To build from source instead, usemake setup(same steps but builds the image locally).

You can run PinkRoosterMcp without cloning the repository by using the pre-built image from Docker Hub.

The 1Password MCP server creates a bridge that allows MCP clients such as Codex and Kiro to manage your 1Password Environments with secure authorization prompts.

This is the 1st, easiest, and cheapest PPT, slides, presentation AI generation MCP Server in the world.

Persistent memory for any AI assistant. Zero token cost until recall. Stores memories in local SQLite, ranks by 6-factor scoring, returns results 79% smaller than JSON. Works with Claude, ChatGPT, Grok, Cursor, Windsurf, and any MCP client.

A MCP server that enables AI assistants to interact with Anki, the spaced repetition flashcard application.

Enables LLM clients to interact with macOS applications through AppleScript. Built using the @beyondbetter/bb-mcp-server library, this server provides safe, controlled execution of predefined scripts with optional support for arbitrary script execution.

An MCP server for WordPress plugin audits

Turn your AI assistant into a digital marketing hub that creates, organizes, and analyzes links and QR Codes on demand.

Connect AI clients to Cal.com scheduling through the Model Context Protocol using the hosted server at mcp.cal.com or a local instance.