skills.rest
telegramdesktop
telegramdesktopOFFICIAL
Explore all skills by this author →
WebsiteGitHub

task-think

Orchestrate multi-phase repository workflows with artifacts under .ai/<task-slug>.

32.6k|7.0k|Updated May 2, 2014
Search Tags:#automation#workflow#codex#artifacts#phases#repo#orchestrate
One-click install
npx skills add https://github.com/telegramdesktop/tdesktop --skill task-think-telegramdesktop
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: task-think
Source: https://github.com/telegramdesktop/tdesktop/tree/main/.codex/skills/task-think
Command: npx skills add https://github.com/telegramdesktop/tdesktop --skill task-think-telegramdesktop

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill orchestrates a multi-phase implementation workflow for a repository using artifact files under .ai/<task-slug> and supports optional fresh codex exec child runs per phase. It enables one prompt to drive context gathering, planning, implementation, verification, and review iterations while keeping the main session context clean.

Core Features & Use Cases

  • Phase-driven workflow: coordinates context, plan, implement, verify, and review phases with clear artifacts.
  • Artifact lifecycle management: automatically creates and maintains .ai/<task-slug> directories and phase logs.
  • Repeatable execution: supports sequential codex exec runs per phase and summarizes results after each phase.
  • Use Case: teams that tackle complex repos with long-running tasks can keep prompts focused per phase while preserving the main session state.

Quick Start

Use the task-think skill to orchestrate a full multi-phase workflow for a repository. The system will create artifacts under .ai/<task-slug>, run phase-specific prompts via codex exec, and generate phase logs for review.

Frequently Asked Questions about task-think

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

FAQPage Schema
How do I orchestrate multi-phase repo tasks end-to-end without losing context?▼

You orchestrate multi-phase repo tasks by running contextual prompts sequentially through context, plan, implement, verify, and review phases. This workflow generates structured artifact files under .ai/<task-slug>, preserving your main session context while driving complex implementations.

What is a phase-driven workflow for repository implementation?▼

A phase-driven workflow structures repository implementation into distinct stages: context gathering, planning, implementation, verification, and review. Each phase generates standardized logs and artifact files, ensuring focused execution and clean context preservation throughout the task lifecycle.

How do I automate context gathering and planning for complex repo tasks?▼

Automate context gathering and planning by executing phase-specific prompts via codex exec. The system automatically creates .ai/<task-slug> directories, runs sequential contextual phases, and generates structured artifact outputs to summarize results after each step.

Can I run codex exec child processes per phase for isolated implementation tasks?▼

Yes, you can run optional fresh codex exec child processes per phase. This allows isolated, repeatable execution of phase-specific prompts while the main workflow orchestrates the overall sequence and maintains the primary session state.

Does this workflow require a YAML frontmatter entry point?▼

Yes, the workflow satisfies the requirement for a YAML frontmatter entry point to initiate the multi-phase task. This standardized entry point drives the phase-driven guidance, artifact lifecycle management, and structured outputs for easy reuse.

What are the limitations of using artifact files for task orchestration?▼

Using artifact files under .ai/<task-slug> for task orchestration requires strict adherence to the artifact lifecycle. If phase logs are not maintained correctly, subsequent contextual prompts may lack the necessary structured outputs, potentially disrupting the sequential implementation workflow.

Related Skills

Explore More Software Engineering Skills→
tneztnez

bootstrap

Initialize the .docent directory structure and configuration for new projects.

Community
Basic
CaseyHaralsonCaseyHaralson
7

orrery-report

Generate structured JSON summaries of AI workflow step statuses, artifacts, and test results.

Community
Basic
Bush2021Bush2021
21

task-think

Orchestrate multi-phase repository implementation workflows with artifact files under .ai/<task-slug>.

Community
Intermediate
goranjovic55goranjovic55

session

Orchestrate session-based task coordination with START, WORK, and END phases.

Community
Advanced