agent-task-manager

Estrellas9

Forks0

Actualizado12 de mayo de 2026 a las 17:30

Comprehensive task management system for AI agents with state tracking and workflow orchestration

Instalación

Instalar con Codex o Claude Copia este prompt, pégalo en Codex, Claude u otro asistente, y deja que revise la página de la skill y la instale por ti.

Fuente

icueth

icueth/devspace-ide-for-claude-code

Abrir repositorio de GitHub Ver repositorios del creador

Descarga

Ocupaciones relacionadasSOC

Basado en la clasificación ocupacional SOC

Desarrolladores de softwareOcupaciones informáticas y matemáticas·SOC 15-1252

SKILL.md

readonly

name	agent-task-manager
description	Comprehensive task management system for AI agents with state tracking and workflow orchestration

Agent Task Manager

This built-in skill provides a robust task management system for AI agents to handle complex workflows, track task states, manage dependencies, and coordinate multi-step processes.

Capabilities

Task Creation: Create tasks with descriptions, priorities, deadlines, and metadata
State Management: Track task states (pending, in_progress, completed, failed, blocked)
Dependencies: Define task dependencies and execution order
Progress Tracking: Monitor task progress with percentage completion and status updates
Error Handling: Handle task failures with retry mechanisms and error reporting
Notifications: Send notifications for task completion, failures, or deadlines
Task Groups: Organize related tasks into groups or projects
Resource Allocation: Manage resource allocation and concurrency limits
History Logging: Maintain complete task execution history for audit trails
Export/Import: Export task data to various formats and import from external sources

Usage Examples

Basic Task Creation

tool: agent-task-manager
action: create_task
task:
  name: "Deploy Application"
  description: "Deploy the latest version to production"
  priority: "high"
  deadline: "2026-03-15T18:00:00Z"
  tags: ["deployment", "production"]

Task Dependencies

tool: agent-task-manager
action: create_task_group
group:
  name: "Feature Implementation"
  tasks:
    - name: "Design API"
      dependencies: []
    - name: "Implement Backend"
      dependencies: ["Design API"]
    - name: "Implement Frontend"
      dependencies: ["Design API"]
    - name: "Integration Testing"
      dependencies: ["Implement Backend", "Implement Frontend"]

Progress Tracking

tool: agent-task-manager
action: update_task
task_id: "task_12345"
updates:
  status: "in_progress"
  progress: 75
  notes: "Backend implementation completed, working on frontend integration"

Security Considerations

Task data is encrypted at rest when sensitive information is involved
Access control ensures only authorized agents can modify tasks
Audit logging tracks all task modifications for security compliance
Resource limits prevent runaway task creation or execution

Configuration

The agent-task-manager skill can be configured with the following parameters:

storage_backend: Storage backend (sqlite, postgres, memory)
max_concurrent_tasks: Maximum number of concurrent tasks (default: 10)
retry_policy: Retry policy for failed tasks (max_retries, backoff_strategy)
notification_channels: Notification channels (email, webhook, internal)
retention_policy: Data retention policy for completed tasks

This skill is essential for any agent that needs to manage complex workflows, coordinate multiple steps, or track progress across extended operations.

Más de este repositorio

mismo repositorio

brand-design-systems

icueth/devspace-ide-for-claude-code

A library of 150 brand and aesthetic design systems (Apple, Stripe, Linear, Vercel, Notion, GitHub, Figma, Spotify, Tesla, and more, plus styles like brutalism, glassmorphism, neumorphism, editorial, minimal). Use when the user wants a UI to look like a brand, match a brand's style or vibe, or adopt a named aesthetic. Read references/<brand>.md for that brand's colors, typography, spacing, radii, and component style before generating any UI.

2026-05-259

brand-guidelines

icueth/devspace-ide-for-claude-code

Apply Anthropic's official brand colors and typography to artifacts for consistent visual identity and professional design standards. A reference for shaping your own.

2026-05-259

canvas-design

icueth/devspace-ide-for-claude-code

Create beautiful visual art in PNG and PDF documents using design philosophy and aesthetic principles for posters, illustrations, and static pieces.

2026-05-259

frontend-design

icueth/devspace-ide-for-claude-code

Frontend design and UI/UX development tools for shipping production-ready interfaces with strong typographic and layout discipline.

2026-05-259

theme-factory

icueth/devspace-ide-for-claude-code

Apply professional font and color themes to artifacts including slides, docs, reports, and HTML landing pages. Ships 10 pre-set themes.

2026-05-259

ui-design

icueth/devspace-ide-for-claude-code

Orchestrates UI and page design. Use when the user asks to design a page, screen, UI, landing page, dashboard, login flow, pricing page, invoice, resume, FAQ, article, slide deck, social card, or any web layout; to make it look like a brand (Apple, Stripe, Linear, Vercel, Notion, ...); or to build a mockup, prototype, wireframe, or HTML preview. Picks a matching layout skill, optionally adopts a brand design system, then writes a single self-contained HTML file to <project>/.devspace/preview/ so DevSpace can preview it.

2026-05-259

Agent Task Manager

This built-in skill provides a robust task management system for AI agents to handle complex workflows, track task states, manage dependencies, and coordinate multi-step processes.

Capabilities

Task Creation: Create tasks with descriptions, priorities, deadlines, and metadata

State Management: Track task states (pending, in_progress, completed, failed, blocked)

Dependencies: Define task dependencies and execution order

Progress Tracking: Monitor task progress with percentage completion and status updates

Error Handling: Handle task failures with retry mechanisms and error reporting

Notifications: Send notifications for task completion, failures, or deadlines

Task Groups: Organize related tasks into groups or projects

Resource Allocation: Manage resource allocation and concurrency limits

History Logging: Maintain complete task execution history for audit trails

Export/Import: Export task data to various formats and import from external sources

Usage Examples

Basic Task Creation

tool: agent-task-manager action: create_task task: name: "Deploy Application" description: "Deploy the latest version to production" priority: "high" deadline: "2026-03-15T18:00:00Z" tags: ["deployment", "production"]

Task Dependencies

tool: agent-task-manager action: create_task_group group: name: "Feature Implementation" tasks: - name: "Design API" dependencies: [] - name: "Implement Backend" dependencies: ["Design API"] - name: "Implement Frontend" dependencies: ["Design API"] - name: "Integration Testing" dependencies: ["Implement Backend", "Implement Frontend"]

Progress Tracking

tool: agent-task-manager action: update_task task_id: "task_12345" updates: status: "in_progress" progress: 75 notes: "Backend implementation completed, working on frontend integration"

Security Considerations

Task data is encrypted at rest when sensitive information is involved

Access control ensures only authorized agents can modify tasks

Audit logging tracks all task modifications for security compliance

Resource limits prevent runaway task creation or execution

Configuration

The agent-task-manager skill can be configured with the following parameters:

storage_backend: Storage backend (sqlite, postgres, memory)

max_concurrent_tasks: Maximum number of concurrent tasks (default: 10)

retry_policy: Retry policy for failed tasks (max_retries, backoff_strategy)

notification_channels: Notification channels (email, webhook, internal)

retention_policy: Data retention policy for completed tasks

This skill is essential for any agent that needs to manage complex workflows, coordinate multiple steps, or track progress across extended operations.