intent-engineering

⛰️ LifeOS — The universal AI Harness designed to move you from Current to Ideal state in both life and work.

The AI harness that moves you from current state to ideal state. An intent engineering platform: it conveys what you ultimately want to your AI on every task. [CODING, BUILDING, CREATING, BUSINESS, LIFE, WORK, ...]

Research and learning toward a future-facing agent system for governing wishes across the Agent Engineering Stack.

立意 Lìyì — 意在笔先 · Persistent, reviewable intent specs for AI-assisted codebases

Diagnostic prompts for understanding AI behavior and intent. Applied robopsychology for modern LLMs and hosted agents.

Pathmode plugin for Claude Code — MCP server + intent-engineering skill pack. Install: /plugin marketplace add pathmodeio/claude-plugin

Pathmode MCP server: deterministic intent preflight before your coding agent builds. Six gates, exact blockers, no model call, keyless. Specs live in intent.md.

Evidence-first map of tools and research for preserving human intent across AI-agent workflows.

Prosthetic cognition architecture for AI agents. Deterministic scaffolding over probabilistic reasoning.

Intent Engineering: the engineering discipline for the human side of AI-led systems | 意图工程:AI 主导系统中面向「人」这一侧的工程学科

Benchmark your harness with head-to-head arena matches, ranked by ELO

Local intent-engineering Journey layer on Codex: confirm Charter, travel by Stations, keep a compact Journal.

Executable research and learning for compiling human goals, constraints, authority, and evidence into agent contracts.

Making slop printable. Prompt-developed prompt-driven CAD

Intent Engineering for Coding Agents: structure, specs, and proof for agentic software engineering. A practical guide for senior devs using coding agents.

A collection of skills for OpenClaw AI agents — context management, intent engineering, and more

Intent-based governance framework for autonomous AI agents. Defining the Golden Path for agentic workflows to ensure business-intent alignment.

The repo-native intent layer for AI-native software engineering — keeps intent and engineering state in git, turns them into workflows, verifies what agents deliver. CLI + MCP.

Sovereign AI+RAG for BIM. A Revit MCP proxy connecting Autodesk Revit MCP Server to AnythingLLM for local LLMs+RAG for deterministic architectural QA/QC and secure document auditing.

Handoffs, Context & Intent Engineering. Compliant to Autonomous Cognitive Entities Framework.