ThomasRohde

Codex Execution Plans (ExecPlans):

This document describes the requirements for an execution plan ("ExecPlan"), a design document that a coding agent can follow to deliver a working feature or system change. Treat the reader as a complete beginner to this repository: they have only the current working tree and the single ExecPlan file you provide. There is no memory of prior plans and no external context.

How to use ExecPlans and PLANS.md

When authoring an executable specification (ExecPlan), follow PLANS.md to the letter. If it is not in your context, refresh your memory by reading the entire PLANS.md file. Be thorough in reading (and re-reading) source material to produce an accurate specification. When creating a spec, start from the skeleton and flesh it out as you do your research.

When implementing an executable specification (ExecPlan), do not prompt the user for "next steps"; simply proceed to the next milestone. Keep all sections up to date, add or split entries in the list at every stoppi

Prompt: Scaffold a Project from `PRD.md`

Use the prompt below with a coding agent to read a PRD.md file and scaffold a project repository, including the recommended files and built-in guidance for how humans should complete any missing details.

Recommended use

Give the agent this prompt in a repository that contains a PRD.md at the root.
The agent should then inspect PRD.md, choose a reasonable project structure, create the starter files, and place clear fill-in guidance where the PRD does not yet specify enough.

CLI-MANIFEST.md — Building Agent-First CLIs

Most CLIs were designed for humans reading terminals. Agents need contracts, not prose.

If output shape changes between commands, errors are only natural language, or writes happen without preview and guardrails, agent automation becomes fragile and expensive.

This manifest defines a practical contract for agent-friendly CLIs:

one structured response envelope for every command
stable error and exit-code semantics
explicit read/write boundaries and safety controls

Rubric for Agent-Performed Assessment of a CLI Implementation (spec-centric)

Goal: Enable an LLM agent to score how “agent-friendly” a CLI is by inspecting the JSON emitted by cli spec (e.g., pubc spec). The rubric emphasizes determinism, non-interactive behavior, machine-readable I/O, discoverability, idempotent semantics, security, streaming/pagination, and portability.

Inputs expected by the agent

The raw JSON from cli spec (primary artifact).

Optionally, --help (text + JSON form), a sample success/error output, and a --version --format json sample.

Outputs produced by the agent

name: agent-creator-engineer description: Use this agent when you need expert assistance with creating high-quality specialized Claude agents following established standards and best practices. This includes analyzing domain requirements, designing agent architectures, implementing comprehensive workflows, and ensuring production-ready agent specifications. The agent specializes in YAML front matter structure, technical documentation patterns, workflow design, and quality assurance standards. Examples:

Context: User wants to create a new specialized agent for a specific technical domain. user: "I need to create an agent for managing Docker containerization and deployment workflows" assistant: "I'll use the agent-creator-engineer agent to design a comprehensive Docker deployment agent following the established CLAUDE.md standards." This requires understanding domain expertise requirements, workflow design, and technical specification patterns for container management.

{
  "mcpServers": {
    "local-api": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "http://localhost:3000/mcp"]
    }
  }
}

	<?xml version="1.0" encoding="UTF-8"?>
	<archimate:model xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:archimate="http://www.archimatetool.com/archimate" name="Archisurance" id="11f5304f" version="5.0.0">
	<folder name="Strategy" id="b93703cf-9575-4d9f-b27f-3423eba81f8a" type="strategy"/>
	<folder name="Business" id="8c90fdfa" type="business">
	<folder name="Processes" id="2966482c">
	<element xsi:type="archimate:BusinessEvent" name="Request for Insurance" id="650"/>
	<element xsi:type="archimate:BusinessEvent" name="Damage Occured" id="651"/>
	<element xsi:type="archimate:BusinessInteraction" name="Formalise Request" id="721"/>
	<element xsi:type="archimate:BusinessInteraction" name="Check and Sign Contract" id="722"/>
	<element xsi:type="archimate:BusinessProcess" name="Inform Customer" id="1302"/>

	<?xml version="1.0" encoding="UTF-8"?>
	<archimate:model xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:archimate="http://www.archimatetool.com/archimate" name="Archi Online Capability Model" id="id-archi-online-capability-model" version="5.0.0">
	<folder name="Strategy" id="id-folder-strategy" type="strategy"/>
	<folder name="Business" id="id-folder-business" type="business">
	<element xsi:type="archimate:BusinessService" name="ArchiMate modeling service" id="id-el-modeling-service">
	<documentation>The user-facing capability for authoring and maintaining ArchiMate models.</documentation>
	</element>
	<element xsi:type="archimate:BusinessService" name="Model review service" id="id-el-review-service">
	<documentation>The user-facing capability for opening a shared model as a read-only review surface.</documentation>
	</element>

	/*
	Author: Thomas Klok Rohde
	Description:
	Insert labels on the current view, or delete labels if no property is selected
	If a view is selected all elements will be labeled, or if an element is selected,
	all view elements with the same type will be labelled.
	Dependencies:
	Thanks to Adam Ernst Bisek: https://github.com/adambisek/string-pixel-width
	History:
	November 16, 2022 : Created