combo

Chain stdin-to-stdout AI tools into a Unix-piped pipeline.

6|1|Updated Jun 23, 2026

Search Tags:#pipeline #sequence #workflows #command-line #ai tool chaining

Ready to run

npx skills add https://github.com/JGalego/Bag-of-Tricks --skill combo

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: combo
Source: https://github.com/JGalego/Bag-of-Tricks/tree/main/combo
Command: npx skills add https://github.com/JGalego/Bag-of-Tricks --skill combo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the need for coordinating multiple AI tools in a single sequence, simplifying complex workflows.

Core Features & Use Cases

Sequence Coordination: Run multiple AI tools in a sequence, with each tool's output automatically passed to the next.
Filter & Analyze: Supports different tool types (filter, analyzer, gate) for various processing steps.
Use Case: Suppose you need to process customer feedback through a sequence of cleaning, summarizing, and flagging steps. This Skill enables you to chain multiple AI tools to efficiently process this feedback in one go.

Quick Start

Create a combo routine using the skills in the Bag-of-Tricks repository: 'combo "frisk | launder | deadpan" < reply.md'.

Frequently Asked Questions about combo

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

FAQPage Schema

How do I chain multiple AI tools together in a single command-line workflow?▼

Chaining AI tools in a command-line workflow requires passing each tool's stdout directly into the next tool's stdin. This approach coordinates complex sequences by automatically feeding output forward through a Unix pipeline.

What is AI tool chaining and when do I need it for complex sequences?▼

AI tool chaining is the process of executing multiple AI tools sequentially in a pipeline. You need it for complex sequences like processing customer feedback through cleaning, summarizing, and flagging steps, where each tool's output becomes the next tool's input.

Do I need stdin to stdout programs to build an AI pipeline sequence?▼

Yes, building an AI pipeline sequence requires each tool to function as a stdin to stdout program. This standard Unix piping mechanism is necessary to compose the workflow and automatically pass output between execution steps.

Can I use Unix piping to coordinate different tool types like filters and analyzers?▼

Yes, you can use Unix piping to coordinate different tool types like filters, analyzers, and gates. This allows you to build varied processing steps within a single workflow sequence to handle tasks like cleaning and summarizing data.

What is the best way to run a sequence of AI tools to process customer feedback?▼

The best way to process customer feedback through a sequence of AI tools is to define a routine using Unix piping. You can chain tools like a filter, summarizer, and flagger in one line to clean, analyze, and flag the input data sequentially.

Are there limitations when using Unix piping for complex AI workflows?▼

The main limitation when using Unix piping for complex AI workflows is that every tool in the sequence must strictly support stdin input and stdout output. Tools that do not adhere to this command-line interface cannot be composed into the pipeline.