Cody Product Builder logo

Cody Product Builder

Organization
ibuildwith-ai
cody-product-builder

Guides knowledge workers and domain experts from idea to finished product through a structured two-phase workflow (Plan and Build). Use this skill whenever the user wants to plan, design, build, or ship a product, app, tool, or feature; add a new version or patch to an existing project; prototype or test an idea quickly; capture a product idea; or types any :cody command. Trigger it even when the user does not say "Cody" but describes wanting to build something, start a project, scope features, write a PRD, or organize product development.

Overview

Publisheribuildwith-ai
Repositorycody-product-builder
Skill namecody-product-builder
Stars
68
Forks
9
Bundled files
27
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 27 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by ibuildwith-ai on GitHub. Read the source before you install it.

Installation

Install the Cody Product Builder AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/ibuildwith-ai/cody-product-builder/tree/main/source/cody-product-builder
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/ibuildwith-ai/cody-product-builder.git /tmp/cody-product-builder
mkdir -p .claude/skills
cp -r /tmp/cody-product-builder/source/cody-product-builder .claude/skills/cody-product-builder
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Cody Product Builder in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Cody Product Builder on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Cody Product Builder is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Cody Product Builder

Agent Instructions

  • Read this entire document (SKILL.md) and commit it to memory.
  • Make sure you check this document often if you need to understand how to process any Cody Product Builder commands.
  • Anything that has [AGENT TODO: to do item] means you need to take action.
  • When you see AGENT ANNOUNCE, anything in between the [message] (tick marks), you will display to the user exactly as stated.
  • After every phase, make sure you re-read this document.
  • {{cfPlaceholders}} used in commands are defined in this document, under the "Template Placeholder Values" section. When you see a placeholder, you will replace it with its appropriate value.
  • The roles are as follows:
    • USER Is the human guiding you in the building process.
    • AGENT That's you! The AI Development AGENT.
  • For detailed phase descriptions, document references, and version naming conventions, see {{cfReferences}}/phases.md.

Template Placeholder Values

These placeholders are a pointer to actual values. They are created here and used throughout various commands. When you encounter a placeholder, you will replace it with its value and consider that as the literal (e.g. {{cfCommands}}/help.md translates to commands/help.md).

The skill-file placeholders ({{cfRoot}}, {{cfAssets}}, {{cfCommands}}, {{cfReferences}}) are paths relative to this skill's own folder.

PlaceholderMaps toDescription
{{cfRoot}}.This skill's root folder.
{{cfAssets}}assetsCody assets folder (templates and other static resources).
{{cfCommands}}commandsCody commands to be executed.
{{cfReferences}}referencesShared reference content loaded on demand by commands.
{{cfProject}}Dynamic -- resolved from cody.jsonProject output folder. Read from cody.json > cody-product-builder > projectPath. Default: cody-projects/product-builder. Resolved on activation and cached for the session. Re-resolved on :cody refresh.
{{cfPlanPhase}}{{cfProject}}/planPlan phase folder.
{{cfWorkPhase}}{{cfProject}}/buildBuild phase folder.
{{cfReleaseNotes}}Dynamic -- resolved from cody.jsonRelease notes folder. Read from cody.json > cody-product-builder > releaseNotesPath. If "{{projectPath}}": resolves to {{cfWorkPhase}}. If "{{projectRoot}}": resolves to project root. Otherwise: resolves to the custom path. Resolved on activation and cached for the session. Re-resolved on :cody refresh.
{{cfPrototypes}}{{cfProject}}/prototypesPrototype output folder. Holds throwaway, self-contained prototype subfolders, each with its own prototype.md and code. Sits outside the Plan and Build phases.
{{cfBestPractices}}{{cfProject}}/best-practicesProject best-practices folder. Holds project-best-practices.md, the project's living record of build learnings. A project-level folder alongside the Plan and Build phases.

File System Checks

  • Always use the placeholder paths (e.g., {{cfPlanPhase}}, {{cfProject}}) when checking for files or folders. Never construct paths manually or use relative paths like ./cody-projects/....
  • Before concluding that a folder is empty or files don't exist, always verify with at least two different methods (e.g., a glob search AND a directory listing). Never make decisions based on a single failed search. This prevents false negatives caused by path resolution issues.

Executing Commands

  • If the USER types any of the commands listed below, read and execute the corresponding command file.
  • Only read command files when they are invoked. Do not pre-load all commands.

:cody help

[AGENT TODO: Read and execute {{cfCommands}}/help.md]

:cody plan

[AGENT TODO: Read and execute {{cfCommands}}/plan.md]

:cody build

[AGENT TODO: Read and execute {{cfCommands}}/build.md]

:cody idea

[AGENT TODO: Read and execute {{cfCommands}}/idea.md]

:cody prototype

[AGENT TODO: Read and execute {{cfCommands}}/prototype.md]

:cody refresh

[AGENT TODO: Read and execute {{cfCommands}}/refresh.md]

Activation

When this skill is activated, AGENT, please execute the following exactly:

  1. Resolve the project configuration. Resolving placeholder paths from cody.json is deterministic work, so the skill ships a script to do it reliably and cheaply.

    Run the resolver script. From the project root, run {{cfRoot}}/scripts/resolve-config.py. It reads cody.json and prints a JSON object: the resolved placeholder paths under resolved (cfProject, cfPlanPhase, cfWorkPhase, cfReleaseNotes, cfPrototypes, cfBestPractices), plus the flags cody_json_exists, has_section, release_notes_path_present, and legacy_project_json_exists. Cache the six resolved values for the rest of the session.

    Fallback -- only if the script cannot run (no Python runtime, or code execution unavailable). Resolve the values by hand:

    • If cody.json exists with a cody-product-builder section: {{cfProject}} = projectPath; {{cfPlanPhase}} = {{cfProject}}/plan; {{cfWorkPhase}} = {{cfProject}}/build; {{cfPrototypes}} = {{cfProject}}/prototypes; {{cfBestPractices}} = {{cfProject}}/best-practices; {{cfReleaseNotes}} from releaseNotesPath ("{{projectPath}}" resolves to {{cfWorkPhase}}, "{{projectRoot}}" to the project root, any other value to that path).
    • If cody.json does not exist: use defaults -- {{cfProject}} = cody-projects/product-builder, with /plan, /build, /prototypes, and /best-practices beneath it, and {{cfReleaseNotes}} = {{cfWorkPhase}}.
    • Cache the resolved values for the rest of the session.

    One-time migration -- missing releaseNotesPath. If cody.json exists with a cody-product-builder section but no releaseNotesPath field (script flag has_section is true and release_notes_path_present is false), handle it here -- the script leaves this to the skill because it needs the USER:

    • Tell the USER where release-notes.md currently lives (check {{cfWorkPhase}}/release-notes.md).
    • Ask: "Would you like to keep release notes there, or move them?" with options:
      1. Keep current location
      2. Project root
      3. Custom path
    • STOP and wait for the USER.
    • If they keep the current location, set releaseNotesPath to "{{projectPath}}" in cody.json.
    • If project root, set releaseNotesPath to "{{projectRoot}}" in cody.json.
    • If custom, ask for the path and set releaseNotesPath to that value in cody.json.
    • If they chose a different location and release-notes.md exists at the current location, move it to the new location.
    • Re-resolve {{cfReleaseNotes}} (re-run the script) and cache it.
  2. Show the USER the following banner (replace {version} with the metadata.version value from this file's frontmatter):

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  Cody Product Builder v{version}
  by iBuildWith.ai
  (c) 2026 Red Pill Blue Pill Studios
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Cody Product Builder is an AI agent skill for domain experts and knowledge workers that takes you from idea to finished product. It provides a structured, repeatable process so your projects are scalable, maintainable, and built right from the start, whether you're starting fresh, improving existing work, or making quick updates in any AI coding environment.

  1. Show the USER a contextual prompt based on their current phase:

    • Check if cody.json exists and has a cody-product-builder section.
      • If it exists and phase is "plan": show "Ready to get started? Type :cody plan to begin, :cody prototype to test an idea first, or :cody help to see all available commands."
      • If it exists and phase is "build": show "What would you like to work on? Type :cody build to continue building, :cody prototype to test an idea, :cody idea to capture a quick thought, :cody refresh to refresh project memory, or :cody help to see all available commands."
      • If it does NOT exist, check if the legacy file cody-projects/product-builder/project.json exists.
        • If the legacy file exists:
          • Tell the USER: "This version of Cody Product Builder uses a new project settings format. Let me migrate your settings."
          • [AGENT TODO: Read and execute {{cfReferences}}/project-settings-check.md]
          • After migration is complete, re-resolve {{cfProject}}, {{cfPlanPhase}}, {{cfWorkPhase}}, and {{cfReleaseNotes}} from the newly created cody.json and cache them for the session.
          • Show the contextual prompt based on the phase from cody.json.
        • If neither file exists: show "What would you like to work on? Type :cody help to see all available commands."
  2. Stop here and wait for the USER.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Cody Product Builder AI skill do?

Guides knowledge workers and domain experts from idea to finished product through a structured two-phase workflow (Plan and Build). Use this skill whenever the user wants to plan, design, build, or ship a product, app, tool, or feature; add a new version or patch to an existing project; prototype or test an idea quickly; capture a product idea; or types any :cody command. Trigger it even when the user does not say "Cody" but describes wanting to build something, start a project, scope features, write a PRD, or organize product development.

Why use Cody Product Builder on TypingMind?

Because you install it once and use it with any model. Cody Product Builder is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Cody Product Builder in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/ibuildwith-ai/cody-product-builder/tree/main/source/cody-product-builder. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Cody Product Builder?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Cody Product Builder?

As many as you like. As long as a model supports skills, you can use Cody Product Builder with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Cody Product Builder AI skill free?

It is published on GitHub by ibuildwith-ai. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇