HN Showcase
Back to browse
GitHub Repository

A portable AI-assisted development workflow. Brainstorm, plan, implement, review, and learn — each cycle makes the next one easier.

19 starsJavaScript

Praxis, my personal take on Compound Engineering with AI

by DFilipeS·Feb 24, 2026·5 points·0 comments
Visit ProjectView on HN

AI Analysis

●●SolidBig BrainShip It

Structured AI workflow beats Every's plugin, but Cursor and Continue solve this already.

Strengths
  • •Context-window-aware design with progressive disclosure avoids token bloat
  • •Extensible subagent architecture lets teams inject custom reviewers for their stack
  • •Tool-agnostic markdown+YAML format works across Amp, Claude Code, and similar platforms
Weaknesses
  • •Requires AI coding agent that supports skills—limits to Amp/Claude Code ecosystem
  • •No evidence of real-world adoption or competitive advantage over monolithic solutions like Cursor
Target Audience

Individual developers and small teams using AI coding agents (Amp, Claude Code)

Similar To

Every Compound Engineering guide · Cursor · Continue IDE

Post Description

Hey HN! I really enjoy Every's approach to Compound Engineering (https://every.to/guides/compound-engineering), but their plugin is tightly tied to their project (Cora) and stack (Ruby/Rails). I also found the files too big, and they used more context window than what I would like for my personal use.

So, with the help of Amp Code CLI, I've built my own take on the compound engineering workflow. I tried to keep it agnostic to project stacks and as efficient as possible, so the context window could be used in the best way. I also wanted it to be extendable (for example, just drop your own subagents for review that are specific to your project). I also wanted to be easy to set up and update, so I made a simple CLI tool that keeps track of files in the `.agents` directory, updates when new versions are found in the repository, and displays a diff in the terminal before overwriting any customisations.

I feel this matches well with my personal preferences when working with AI agents, but I would love to have feedback from more people.

Similar Projects

AI/ML●●●Banger

YantrikDB – persistent memory for AI agents

Bundled 7MB embedder means zero network calls or model downloads for agent memory.

Zero to OneWizardrySolve My Problem
pranabsarkar
1123d ago
Developer Tools●●Solid

ai11y – A structured UI context layer for AI agents

The describe → plan → act split is an elegant, accessibility-inspired way to give LLM agents actionable UI context: annotate with data-ai-* attributes or use the Marker component, call describe(), send it to a planner, then client.act() executes DOM instructions. It's a clever middle layer that turns messy DOM state into structured inputs for server-side planning, though adoption will hinge on robust selector semantics and out-of-the-box integrations with popular LLMs and automation backends.

Big BrainNiche Gem
maerzhase3000
113mo ago
Developer Tools●●Solid

Agent Audit Kit v0.1 – deterministic replay + stress for LLM agents

Deterministic capture + replay for LLM agents is a practical, under-served problem and this repo actually ships a 'golden run' zip with cold‑run verification hashes — that’s the kind of evidence chain auditors want. The focus on portable evidence bundles and stress verification suggests useful forensics and load testing of agent logic, but the release page looks early-stage; I'd like to see integrations (tooling for popular agent frameworks), richer docs, and example pipelines before I'd evangelize it.

Niche GemSolve My ProblemShip It
helpfuldolphin
103mo ago