How do I install the Ontology skill?

Run npx skillmds@latest add q00/ontology in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Ontology safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Ontology?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Ontology free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Ontology?

Q00 (@q00) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Ontology by q00 — Agent Skill

Name: Ontology
Author: Q00

Ontology

Add a semantic boundary layer on top of a closed Seed. Use Seed.goal as idea, ask for boundary and exactly three properties, and replace the top-level ## Ontology section in .symposium/scratch/socrates.md. Triggers (EN): ontology, create ontology, define semantic boundary, add meaning boundary Do NOT use when: no Seed exists yet (-> socrates), Seed content must evolve (-> evolve-step), or the full convergence harness is needed (-> interview-harness) vs evolve-step: Evolve-step changes the Seed; Ontology adds a boundary layer above the Seed.

Q00 b1d1665 3 files · 4.0 KB Updated Sep 17, 2026

File contents

Ontology

Details

Ontology defines what the Seed's goal means and where that meaning stops. It does not ask for actions; acceptance_criteria already covers actions.

Input

Read the latest Seed from .symposium/scratch/socrates.md top-level ## Seed. If there is no Seed, stop and ask the user to run socrates first.

Ontology Shape

ontology:
  idea: <Seed.goal exactly>
  boundary: <what is inside and outside the idea>
  properties:
    - <property 1>
    - <property 2>
    - <property 3>

Process

Copy Seed.goal exactly into ontology.idea.
Ask one boundary question: "Where does this idea's meaning begin and end?"
Ask for exactly three properties inside that boundary.
Replace the top-level ## Ontology section in .symposium/scratch/socrates.md.

Use request_user_input in Codex Plan mode when available, AskUserQuestion in Claude when available, or normal conversational questions otherwise.

Question Style

Prefer multiple-choice prompts for both boundary and properties, but keep them editable. Choices are candidate boundaries/properties, not final ontology terms.

Boundary prompt shape:

Where does this idea's meaning begin and end?
A. <boundary candidate>
B. <boundary candidate>
C. <boundary candidate>
D. Other / define the boundary in your own words

Properties prompt shape:

Which three properties must be inside that boundary?
A. <property>, <property>, <property>
B. <property>, <property>, <property>
C. <property>, <property>, <property>
D. Other / list exactly three properties

Only write terms the user selected or wrote.

Rules

idea and boundary are different.
Every property must be inside the boundary.
Exactly three properties are required.
Do not ask for data types.
Do not ask for actions.
Only user-provided boundary and property terms enter the YAML.

Convergence Comparison

The interview-harness compares the previous ontology and the current ontology:

idea: 1 if identical, otherwise 0.
boundary: 1 only if the user confirms the two boundaries mean the same thing.
properties: Jaccard overlap between the two property sets.

Average the three scores. A score of 0.85 or higher means converged.

---name: ontologydescription: Add a semantic boundary layer on top of a closed Seed. Use Seed.goal as idea, ask for boundary and exactly three properties, and replace the top-level ## Ontology section in .symposium/scratch/socrates.md. Triggers (EN): ontology, create ontology, define semantic boundary, add meaning boundary Do NOT use when: no Seed exists yet (-> socrates), Seed content must evolve (-> evolve-step), or the full convergence harness is needed (-> interview-harness) vs evolve-step: Evolve-step changes the Seed; Ontology adds a boundary layer above the Seed.---# Ontology## DetailsOntology defines what the Seed's goal means and where that meaning stops. It does not ask for actions; `acceptance_criteria` already covers actions.## InputRead the latest Seed from `.symposium/scratch/socrates.md` top-level `## Seed`. If there is no Seed, stop and ask the user to run `socrates` first.## Ontology Shape```yamlontology: idea: <Seed.goal exactly> boundary: <what is inside and outside the idea> properties: - <property 1> - <property 2> - <property 3>```## Process1. Copy `Seed.goal` exactly into `ontology.idea`.2. Ask one boundary question: "Where does this idea's meaning begin and end?"3. Ask for exactly three properties inside that boundary.4. Replace the top-level `## Ontology` section in `.symposium/scratch/socrates.md`.Use `request_user_input` in Codex Plan mode when available, `AskUserQuestion` in Claude when available, or normal conversational questions otherwise.## Question StylePrefer multiple-choice prompts for both boundary and properties, but keep them editable. Choices are candidate boundaries/properties, not final ontology terms.Boundary prompt shape:```markdownWhere does this idea's meaning begin and end?A. <boundary candidate>B. <boundary candidate>C. <boundary candidate>D. Other / define the boundary in your own words```Properties prompt shape:```markdownWhich three properties must be inside that boundary?A. <property>, <property>, <property>B. <property>, <property>, <property>C. <property>, <property>, <property>D. Other / list exactly three properties```Only write terms the user selected or wrote.## Rules- `idea` and `boundary` are different.- Every property must be inside the boundary.- Exactly three properties are required.- Do not ask for data types.- Do not ask for actions.- Only user-provided boundary and property terms enter the YAML.## Convergence ComparisonThe `interview-harness` compares the previous ontology and the current ontology:- idea: 1 if identical, otherwise 0.- boundary: 1 only if the user confirms the two boundaries mean the same thing.- properties: Jaccard overlap between the two property sets.Average the three scores. A score of 0.85 or higher means converged.

Frequently asked questions

Ontology

File contents

Ontology

Details

Input

Ontology Shape

Process

Question Style

Rules

Convergence Comparison

Related skills

Frequently asked questions

How do I install the Ontology skill?

What does the Ontology skill do?

Is Ontology safe to use?

Which AI agents work with Ontology?

Is Ontology free to use?

Who published Ontology?