complete-task

Name: complete-task
Availability: InStock
Author: Accomplish

Enforce explicit task completion via complete_task with status and summaries.

10.9k|1.3k|Updated Jan 14, 2026

Search Tags:#MCP #task-completion #status-reporting #AI-agent #complete-task #workflow-control

Ready to run

npx skills add https://github.com/accomplish-ai/accomplish --skill complete-task

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: complete-task
Source: https://github.com/accomplish-ai/accomplish/tree/main/packages/agent-core/mcp-tools/complete-task
Command: npx skills add https://github.com/accomplish-ai/accomplish --skill complete-task

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This tool enforces a formal completion signal to the agent, ensuring tasks don’t linger unfinished or ambiguous.

Core Features & Use Cases

Enforces explicit task completion by requiring the agent to call complete_task when a task finishes or encounters a blocker.
Provides structured statuses (success, blocked, partial) and preserves the original request context for review.
Supports clear summaries of what was accomplished and what remains, enabling reliable downstream automation.

Quick Start

Use the complete_task tool to report a successful completion, e.g., call complete_task with status: "success", original_request_summary: "Summarize the user's request", summary: "All steps completed", remaining_work: "" (optional).

Frequently Asked Questions about complete-task

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I ensure task completion signals are explicit and tracked in my workflow?▼

Explicit task completion requires calling complete_task with a status (success, blocked, or partial), original request summary, and work summary. This enforces a formal signal so tasks don't remain ambiguous or unfinished, enabling reliable downstream automation.

What statuses can I report when completing a task?▼

Task completion supports three statuses: success for fully completed work, blocked when work cannot proceed, and partial for incomplete progress. Each status preserves the original request context and requires a concise summary of what was accomplished.

Can I document remaining work when a task is incomplete?▼

Yes, the complete_task tool includes an optional remaining_work field to document what still needs completion. This enables clear handoffs and helps downstream processes understand what blocked or partial tasks require next.

Does task completion tracking work within MCP runtime without external setup?▼

Task completion runs natively within the MCP runtime without external installations or dependencies. The complete_task tool is available immediately for structuring task endpoints across all lifecycle stages.

How do I preserve context when reporting task status?▼

Complete_task captures the original_request_summary to preserve what was initially asked, alongside a summary of what was done. This context retention ensures reviewers and downstream automation understand intent and outcomes.