demineur4356435

demineur4356435 / ts-planner-agent

Public

Task planning agent for LLMs — hierarchical plans, DAG dependencies, plan adaptation, Mermaid/Markdown export. TS, Express, MIT.

agent ai ai-agent anthropic goal-decomposition
20
0
100% credibility
Found Apr 19, 2026 at 20 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

A TypeScript library that uses AI to decompose complex goals into ordered, dependency-aware step-by-step plans with effort estimates, risks, and adaptation capabilities.

How It Works

1
🔍 Find the smart planner

You stumble upon a handy tool online that uses AI to turn your big dreams or tough tasks into easy-to-follow step-by-step plans.

2
📥 Set it up on your computer

You bring the tool home to your machine and prepare it so it's ready to help you right away.

3
🔗 Link to an AI helper

You connect it to a thinking AI service, like a smart brain, so it can understand and plan deeply.

4
💭 Describe your big goal

You simply tell it what you want to achieve, like planning a website or learning a skill.

5
✨ Get your perfect plan

It instantly creates a detailed roadmap with ordered steps, time guesses, risks, and what needs to happen first.

6
📊 View and share the plan

You see the plan as a neat list, colorful chart, or timeline, perfect for printing or following along.

7
🔄 Tweak if things change

If a step hits a snag, you explain what went wrong and it smartly updates the plan for you.

🎉 Tackle anything!

Now you have a clear, confident path to finish your project, feeling organized and excited.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 20 to 20 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is ts-planner-agent?

ts-planner-agent is a TypeScript library that uses LLMs to turn vague goals into structured, dependency-aware task plans—like a github task planner on steroids. Feed it a high-level objective, like "ship a CRUD API," and it outputs a DAG of steps with effort estimates, risks, and verification checks, plus exports to Mermaid diagrams, Markdown task lists, or Gantt charts. Built with Express for a REST API and supporting local models via Ollama, it handles plan adaptation when steps fail, making it a solid task planning agent for dynamic workflows.

Why is it gaining traction?

It stands out with hierarchical decomposition, cycle-safe DAG dependencies, and real-time adaptation—features rare in basic task planning tools. Developers hook it via simple CLI commands, API endpoints like POST /plan or /adapt, or direct imports, integrating seamlessly with task github actions or github task copilot setups. Exports to github task list markdown or Mermaid make sharing plans effortless, while multi-provider support (OpenAI, Anthropic, Groq) keeps costs low.

Who should use this?

DevOps engineers building runbooks, researchers outlining experiments, or indie devs plotting project roadmaps will find it handy for github task flow and task planning sheets. It's perfect for teams needing task planning examples in robotics sims, safety talks, or learning paths, especially if you're chaining it with github task scheduler automation.

Verdict

With 20 stars and a 1.0% credibility score, it's early-stage but boasts solid docs, Docker support, and mock-tested coverage—worth a spin for task planning toolbox experiments. Try the CLI for quick wins if you need lightweight, LLM-driven planning without heavy setup.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.