skills.rest
qqlcx5
qqlcx5COMMUNITY
Explore all skills by this author →
WebsiteGitHub

ontology

Create, query, and validate typed knowledge graphs stored as JSONL entity records.

Updated Aug 15, 2026
Search Tags:#knowledge graph#ontology#schema validation#agent memory#jsonl#graph traversal#entity relations
One-click install
npx skills add https://github.com/qqlcx5/skills-hub --skill ontology-qqlcx5
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ontology
Source: https://github.com/qqlcx5/skills-hub/tree/main/skills/ontology
Command: npx skills add https://github.com/qqlcx5/skills-hub --skill ontology-qqlcx5

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, and includes scripts (resource) and references (resource) components.

What problem does it solve? Agents lose track of structured information across sessions and skills. This Skill provides a typed knowledge graph where entities like people, projects, tasks, and events are stored with validated relations, giving agents persistent, queryable memory that multiple skills can share. ## Core Features & Use Cases - Typed Entity CRUD: Create, update, query, and delete entities (Person, Project, Task, Event, Document, and more) stored in an append-only JSONL graph file. - Relation Management & Traversal: Link entities with typed relations, query dependencies in both directions, and detect circular task blockers. - Schema Constraint Validation: Enforce required properties, enums, forbidden fields, cardinality rules, and acyclicity through a YAML schema. - Use Case: Ask the agent to remember that Alice owns the Website Redesign project, then later query all open tasks blocking that project and get a validated, structured answer. ## Quick Start Ask the agent to remember a new person named Alice and link her as the owner of a project called Website Redesign in the knowledge graph.

Frequently Asked Questions about ontology

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a knowledge graph for agent memory in Python?▼

Use the ontology.py CLI to create typed entities and relations stored in a JSONL graph file. Run commands like create, relate, and query to build the graph incrementally, with each operation appended as a validated record.

How to query relationships between tasks and projects?▼

Use the related command with an entity ID and relation type to traverse the graph in outgoing, incoming, or both directions. For example, query all tasks of a project via the has_task relation or find blockers with incoming blocks relations.

Can the knowledge graph detect circular task dependencies?▼

Yes. Relations marked acyclic: true in the schema are validated with depth-first search during the validate command. Any cycle in relations like blocks or depends_on is reported as a validation error.

What storage format does the ontology graph use?▼

The graph is stored as an append-only JSONL file at memory/ontology/graph.jsonl, where each line is a create, update, delete, relate, or unrelate operation. For complex graphs, the documentation suggests migrating to SQLite.

What are the limitations of JSONL-based graph storage?▼

Every query reloads and replays the entire operation log, so performance degrades on large graphs. Complex multi-hop traversals and aggregations require custom Python code rather than built-in query language support.

Related Skills

Explore More Software Engineering Skills→
eddowdingeddowding

Extract Entities

Extract people and organizations from emails into Obsidian CRM notes.

Community
Advanced
JansenAnalyticsJansenAnalytics
5

project-planner

Decompose project goals into actionable tasks with dependencies and sub-agent scheduling.

Community
Advanced
semantica-agisemantica-agi
3.0k

temporal

Apply as_of constraints to temporal graph queries, snapshots, and causal traces.

Official
Advanced
codatacodata
8

neo4j_expert

Convert Croissant dataset JSON-LD files into a Neo4j knowledge graph.

Official
Intermediate