skills-graph 0.1.2

Persist, retrieve and evolve AI skills in Memgraph

skills-graph

A small library to persist, retrieve and evolve AI skills in Memgraph.

Graph Model

(:Skill {name, description, content, created_at, updated_at})
(:Skill)-[:DEPENDS_ON]->(:Skill)
(:Session)-[:USED_SKILL]->(:Skill)

Quick Start

from skills_graph import SkillGraph, Skill

# Connect (uses MEMGRAPH_URL, MEMGRAPH_USER, MEMGRAPH_PASSWORD env vars by default)
sg = SkillGraph()

# Prepare the database schema (constraints + indexes)
sg.setup()

# Store a skill
sg.add_skill(Skill(
    name="memgraph-cypher",
    description="Writing Cypher queries for Memgraph",
    content="# Cypher for Memgraph\n\nUse MATCH, CREATE, MERGE ...",
))

# Retrieve by name
skill = sg.get_skill("memgraph-cypher")

# Search
sg.search_by_name("memgraph")

# Dependencies
sg.add_dependency("advanced-cypher", "memgraph-cypher")
deps = sg.get_dependencies("advanced-cypher")

# List all
all_skills = sg.list_skills()

# Update
sg.update_skill("memgraph-cypher", content="updated content")

# Delete
sg.delete_skill("memgraph-cypher")

Codex Hook Integration

When used through agent-context-graph's Codex hook adapter, SkillGraphConnector records direct tool names like get_skill and Codex MCP-style names like mcp__skills__get_skill.

It also records local reads of valid Agent Skills definitions, such as /skills/<name>/SKILL.md, as skill usage. This supports runtimes where using a skill appears as a filesystem or shell tool reading the skill's SKILL.md file rather than as a dedicated get_skill tool call.

Search/list tool results can be Python lists or JSON tool content containing skill objects with name fields. Today those results are recorded through USED_SKILL with result actions; future work will split weaker surfacing signals into a separate relationship.

Installation

uv sync

Testing

uv run pytest tests/ -v