Ghm Sot Builder logo

Ghm Sot Builder

Community
mattgierhart
ghm-sot-builder

Creates new Source of Truth (SoT) files when existing templates don't fit your needs. Triggers on requests to create a new SoT file, add a new artifact type, or when user says "I need to track [X] but there's no SoT for it", "create SoT", "new source of truth". Outputs a properly structured SoT.*.md file with ID prefix, frontmatter, and update protocol.

Overview

Publishermattgierhart
RepositoryPRD-driven-context-engineering
Skill nameghm-sot-builder
Stars
179
Forks
11
Bundled files
3
LicenseMIT
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.

  • 3 bundled files

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

  • Open source

    Published by mattgierhart on GitHub. Read the source before you install it.

Installation

Install the Ghm Sot 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.
  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/mattgierhart/PRD-driven-context-engineering.git /tmp/PRD-driven-context-engineering
mkdir -p .claude/skills
cp -r /tmp/PRD-driven-context-engineering/plugins/prd-ce/skills/ghm-sot-builder .claude/skills/ghm-sot-builder
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Ghm Sot 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 Ghm Sot 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 Ghm Sot 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.

SoT Builder

Create new Source of Truth files that fit your product's unique needs while maintaining template purity.

When to Use This Skill

This skill is for rare occasions (3-4 times per product lifecycle) when:

  • You fork this repo and the existing SoT files don't cover your artifact types
  • Your product needs to track a concept that doesn't fit existing ID prefixes
  • You need to consolidate scattered documentation into a canonical SoT

Do NOT use when:

  • An existing SoT file covers your need (just add entries there)
  • You want to track temporary/session-specific data (use temp/ instead)
  • The artifact type is already covered by ghm-id-register

Workflow Overview

  1. Identify Need → Confirm no existing SoT fits
  2. Design Schema → Define ID prefix, categories, required fields
  3. Draft Template → Create pure structure (no methodology teaching)
  4. Validate Purity → Apply the litmus test
  5. Register & Integrate → Update SoT.README.md and ID system

Core Output Template

See assets/sot-template.md for the copy-paste starter template.

ElementDefinitionRequired
YAML Frontmatterversion, purpose, id_prefix, authorityYes
Purpose BlockSingle paragraph explaining what this tracksYes
Navigation SectionCategory links for quick accessYes
Entry TemplateRepeatable structure for each IDYes
Cross-Reference IndexBidirectional links to other SoT filesYes
Update ProtocolWhen/how to add new entriesYes

Step 1: Identify the Need

Before creating a new SoT, verify:

Checklist

  • Searched existing SoT files for similar artifact types
  • Confirmed the concept is durable (survives multiple sessions)
  • Confirmed the concept needs unique IDs for cross-referencing
  • Confirmed this isn't better served by a subsection in an existing SoT

Questions to Answer

  1. What artifact type does this track?

    • Bad: "Notes" (too generic)
    • Good: "Partner Integration Contracts" (specific, durable)
  2. Why can't existing SoT files handle this?

    • Valid: "We need different fields than API contracts for partner integrations"
    • Invalid: "I just want a separate file" (preference, not need)
  3. What ID prefix will you use?

    • Must be unique across the system
    • 2-4 uppercase letters
    • Check SoT/SoT.UNIQUE_ID_SYSTEM.md for existing prefixes

Step 2: Design the Schema

Define the structure before writing:

ID Prefix Selection

Format: [PREFIX]-[XXX]
Examples: PIC-001, INT-042, MIG-003

Rules:

  • 2-4 uppercase letters
  • Descriptive but short
  • Unique across all SoT files
  • Check existing prefixes in SoT/SoT.UNIQUE_ID_SYSTEM.md

Category Ranges

Reserve ID ranges for logical groupings:

markdown
**Category A** (XXX-001 to XXX-099):
**Category B** (XXX-101 to XXX-199):
**Category C** (XXX-201 to XXX-299):

Required Fields per Entry

Define the minimum fields every entry needs:

FieldPurposeExample
IDUnique identifierPIC-001
TitleHuman-readable name"Stripe Payment Integration"
StatusCurrent stateActive / Deprecated / Planned
CreatedOrigin date2025-01-10
Last UpdatedLast modification2025-01-15
Related IDsCross-referencesBR-101, API-045

Optional Fields

Add domain-specific fields as needed, but keep them structural (not methodology):

  • Good: "Partner Contact" (data field)
  • Bad: "Best Practices for Partner Selection" (methodology teaching)

Step 3: Draft the Template

Use assets/sot-template.md as your starting point.

Structure Sections

  1. YAML Frontmatter - Metadata for the file
  2. Title & Purpose Block - What this SoT tracks
  3. Navigation by Category - Quick links to entries
  4. Entry Template - Repeatable structure (copy for each new entry)
  5. Deprecated Section - Where old entries go
  6. Cross-Reference Index - Bidirectional links
  7. Update Protocol - When/how to maintain

Template Purity Rules

KEEP in the template (self-documentation):

  • When to add new entries
  • Required fields checklist
  • Cross-reference integrity checks
  • File-specific maintenance procedures

MOVE to skill references (methodology teaching):

  • "Best practices" for this domain
  • "What makes a good entry"
  • Example workflows beyond format
  • Evaluation criteria

Litmus Test

For each section, ask:

"Is this teaching me how to maintain the FILE STRUCTURE, or teaching me DOMAIN KNOWLEDGE about what makes good content?"

  • File structure maintenance → Keep in template
  • Domain knowledge → Move to skill references

Step 4: Validate Purity

Before finalizing, run this checklist:

Purity Checklist

  • No "how to analyze" or "how to evaluate" content
  • No "best practices" or "key learnings" sections
  • No cross-file workflows (multi-file coordination)
  • No "what makes a good entry" evaluation criteria
  • Examples show FORMAT only, not instructional CONTENT
  • Self-documentation is under 20% of total file size

Self-Documentation Checklist

  • Has "Update Protocol" section
  • Documents when to add new IDs
  • Includes cross-reference integrity checks
  • Specifies required fields for new entries
  • Self-documentation is file-specific (not generic)

Context Efficiency Checklist

  • Template can be read without loading other files
  • Methodology examples are in skill references, not template
  • General governance references SoT.README.md

Step 5: Register and Integrate

After creating the SoT file:

Update SoT.README.md

Add entry to the structure table:

markdown
| `SoT.{YOUR_FILE}.md` | {PREFIX}-### | {Purpose description} |

Update SoT.UNIQUE_ID_SYSTEM.md

Add new prefix to Section 1.2 Standard Prefixes:

markdown
| **{PREFIX}** | {Meaning} | `SoT.{YOUR_FILE}.md` |

Create Index Tables

Add empty index table in Part 2 of SoT.UNIQUE_ID_SYSTEM.md:

markdown
#### {Your Type} ({PREFIX}-XXX)

| ID | Title | Status | Used By |
|----|-------|--------|---------|
| {PREFIX}-001 | {Title} | Active | {IDs} |

Quality Gates

Pass Checklist

  • ID prefix is unique across all SoT files
  • Template follows purity standard
  • Update protocol included
  • Cross-reference index structure defined
  • SoT.README.md updated
  • SoT.UNIQUE_ID_SYSTEM.md updated

Testability Check

  • Can create a new entry using only the template instructions
  • Cross-references can be validated programmatically
  • Entries can be found by ID search

Anti-Patterns

PatternExampleFix
Methodology in template"Best practices for partner selection"Move to skill references
Duplicate prefixUsing BR- for a new fileChoose unique prefix
Too generic"Notes.md"Be specific: "Partner_Integrations.md"
No update protocolTemplate with no maintenance sectionAdd "Update Protocol" section
Orphan SoTNot registered in SoT.README.mdAlways register new files

Bundled Resources

  • references/sot-patterns.md — Common patterns across existing SoT files
  • references/examples.md — Before/after examples of SoT creation
  • assets/sot-template.md — Copy-paste starter template

Handoff

After creating a new SoT:

  1. SoT file exists at SoT/SoT.{NAME}.md
  2. SoT.README.md lists the new file
  3. SoT.UNIQUE_ID_SYSTEM.md has the new prefix
  4. Ready to use ghm-id-register for adding entries

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 Ghm Sot Builder AI skill do?

Creates new Source of Truth (SoT) files when existing templates don't fit your needs. Triggers on requests to create a new SoT file, add a new artifact type, or when user says "I need to track [X] but there's no SoT for it", "create SoT", "new source of truth". Outputs a properly structured SoT.*.md file with ID prefix, frontmatter, and update protocol.

Why use Ghm Sot Builder on TypingMind?

Because you install it once and use it with any model. Ghm Sot 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 Ghm Sot Builder in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/mattgierhart/PRD-driven-context-engineering/tree/main/plugins/prd-ce/skills/ghm-sot-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 Ghm Sot 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 Ghm Sot Builder?

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

Is the Ghm Sot Builder AI skill free?

Yes. It is published on GitHub by mattgierhart under the MIT license. 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 👇