Pu.sh – a full coding-agent harness in 400 lines of shell
AgentI originally was just messing with pi-autoresearch. Gave it a sample task to build the most portable coding agent.First cut was 6 KB of shell. Great for one-shots, unusable interactively. I was shocked it actually worked.Started building up -- adding features — but with a self-imposed rule: no new d
- Best for
- shell-based coding agent orchestration, task execution via shell commands, customizable coding workflows
- Type
- Agent
- Score
- 37/100
- Best alternative
- LangChain
Capabilities3 decomposed
shell-based coding agent orchestration
Medium confidencePu.sh implements a lightweight coding agent framework entirely in shell scripting, allowing for seamless orchestration of coding tasks through simple command-line interactions. It leverages shell functions and scripts to manage the execution of coding tasks, making it highly portable and easy to integrate into existing shell environments. This approach minimizes dependencies and keeps the entire framework concise, enabling rapid deployment and iteration.
The entire framework is implemented in just 400 lines of shell code, making it extremely lightweight and easy to modify for specific tasks.
More lightweight and easier to modify than larger frameworks like Ansible or Terraform, which require more complex setups.
task execution via shell commands
Medium confidencePu.sh allows users to define and execute coding tasks directly through shell commands, utilizing built-in functions to handle various programming operations. This capability enables users to run scripts, compile code, or execute tests with simple command-line instructions. The use of shell commands provides a familiar interface for developers, making it easy to integrate into their existing workflows without the need for additional tooling.
Directly integrates task execution into the shell environment, allowing for immediate feedback and interaction.
More straightforward for shell users compared to GUI-based tools that abstract command execution.
customizable coding workflows
Medium confidencePu.sh enables users to create customizable workflows by defining their own shell functions and scripts, allowing for tailored coding processes that fit specific project needs. This flexibility is achieved through a modular design where users can easily add or modify functions to extend the capabilities of the coding agent. The ability to customize workflows directly in shell scripts provides a high degree of control and adaptability.
The ability to define custom workflows in shell scripts allows for a high level of personalization and adaptability not found in rigid frameworks.
More flexible than traditional IDEs that impose strict workflows and processes.
Capabilities are decomposed by AI analysis. Each maps to specific user intents and improves with match feedback.
Related Artifactssharing capabilities
Artifacts that share capabilities with Pu.sh – a full coding-agent harness in 400 lines of shell, ranked by overlap. Discovered automatically through the match graph.
paseo
Orchestrate coding agents remotely from your phone, desktop and CLI
Multi – Frontier AI Coding Agent
Frontier AI Coding Agent for Builders Who Ship.
BLACKBOXAI #1 AI Coding Agent and Coding Copilot
BLACKBOX AI is an AI coding assistant that helps developers by providing real-time code completion, documentation, and debugging suggestions. BLACKBOX AI is also integrated with a variety of developer tools such as Github Gitlab among others, making it easy to use within your existing workflow.
GoCodeo
An AI Coding & Testing Agent.
Omar – A TUI for managing 100 coding agents
We were both genuinely impressed by Claude Code after it helped each of us fix nasty CI problems overnight. Doing those fixes manually would have taken days.After that experience, we each found ourselves struggling through Ctrl+Tab through multiple Claude Code windows in our terminals. While we enjo
Agent-of-empires: OpenCode and Claude Code session manager
Hi! I’m Nathan: an ML Engineer at Mozilla.ai: I built agent-of-empires (aoe): a CLI application to help you manage all of your running Claude Code/Opencode sessions and know when they are waiting for you.- Written in rust and relies on tmux for security and reliability - Monitors state of cli s
Best For
- ✓developers looking for a minimalistic coding agent solution
- ✓developers who prefer command-line interfaces for automation
- ✓developers looking for flexibility in their coding processes
Known Limitations
- ⚠Limited to shell environments, may not support complex programming tasks requiring advanced tooling
- ⚠Performance may vary based on shell execution speed
- ⚠Limited error handling compared to more robust programming environments
- ⚠Shell command execution may vary based on the user's shell configuration
- ⚠Requires knowledge of shell scripting to fully leverage customization
- ⚠Complex workflows may become difficult to manage without proper organization
Requirements
Input / Output
UnfragileRank
UnfragileRank is computed from adoption signals, documentation quality, ecosystem connectivity, match graph feedback, and freshness. No artifact can pay for a higher rank.
About
Show HN: Pu.sh – a full coding-agent harness in 400 lines of shell
Categories
Alternatives to Pu.sh – a full coding-agent harness in 400 lines of shell
OpenAI's official agent framework — agents, handoffs, guardrails, sessions, built-in tracing.
Compare →Anthropic's official agent SDK — the Claude Code harness (tools, MCP, subagents, permissions) as a library.
Compare →Most-starred open-source browser-agent library — agents drive real browsers via Playwright + any LLM.
Compare →Are you the builder of Pu.sh – a full coding-agent harness in 400 lines of shell?
Claim this artifact to get a verified badge, access match analytics, see which intents users search for, and manage your listing.
Get the weekly brief
New tools, rising stars, and what's actually worth your time. No spam.
Data Sources
Looking for something else?
Search →