Manages complex repository tasks by creating project artifacts and driving multi-phase execution through a single prompt.

Install

mkdir -p .claude/skills/task-think && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/3711" && unzip -o skill.zip -d .claude/skills/task-think && rm skill.zip

Installs to .claude/skills/task-think

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Orchestrate a multi-phase Telegram Desktop implementation workflow with persistent per-project task artifacts under .ai. Use when Codex should drive context gathering, planning, plan assessment, implementation, Debug build verification, review, and native-Windows text normalization through bounded phase handoffs while keeping the parent task lean. Uses current spawn_agent, wait_agent, send_message, followup_task, and interrupt_agent semantics.
447 chars✓ has a “when” triggerlonger than Claude Code's old 250-char listing cap (fine on current versions)
Advanced

Key capabilities

  • Orchestrate multi-phase implementation workflows
  • Maintain persistent project artifacts in .ai/ directories
  • Delegate tasks to subagents for specific phases
  • Perform Windows text normalization
  • Verify builds and manage review iterations

How it works

It manages a project-based pipeline where each task is broken into phases (context, planning, implementation, verification, review) with artifacts stored in a structured .ai/ directory.

Inputs & outputs

You give it
Task description and optional constraints
You get back
Implemented feature with verified artifacts

When to use task-think

  • Implementing multi-phase features
  • Automating complex repository refactoring
  • Managing multi-step dev workflows
  • Task-driven coding assistance

About task-think

Uses a task-slug based system to store workflow progress and artifacts in .ai/ directories. It triggers recursive executions to handle context gathering, planning, and implementation phases.

Orchestrate a multi-phase implementation workflow for this repository with artifact files under .ai/task-slug and optional fresh codex exec child runs per phase. Use when the user wants one prompt to drive context gathering, planning, implementation, verification, and review iterations while keeping the main session context clean.

When not to use it

  • For simple, single-step tasks that do not require planning
  • When the environment does not support subagent delegation

Limitations

  • Requires subagent support for optimal performance
  • Build verification is limited to the current environment

How it compares

It provides a formal, artifact-driven orchestration layer that keeps the main session context clean by delegating work to subagents.

Compared to similar skills

task-think side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
task-think (this skill)11moNo flagsAdvanced
feature-design-assistant67moNo flagsIntermediate
github-project-management46moReviewAdvanced
create-plans18moReviewIntermediate

Try saying

Example prompts that trigger this skill in your AI assistant.

Search skills

Search the agent skills registry