메뉴
Store and retrieve knowledge in the Alhazen TypeDB knowledge graph - remember, recall, organize papers and notes
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
TypeDB-backed ontological memory with schema-driven retrieval. Introspects the live knowledge graph schema, composes TypeQL queries dynamically, and combines graph traversal with embedding-based semantic search for three-stage retrieval (plan, execute, organize with provenance).
REQUIRED FIRST — starts TypeDB + dashboard via Docker, loads the Alhazen base schema. Must install before any other Alhazen skill.
Brief description of what this skill does
Design and implement TypeDB-backed curation skills using the Skillful Alhazen methodology
Search the web using SearXNG (self-hosted metasearch - no API key needed)
| name | typedb-notebook |
| description | Store and retrieve knowledge in the Alhazen TypeDB knowledge graph - remember, recall, organize papers and notes |
Use this skill to store and retrieve knowledge in the Alhazen TypeDB knowledge graph. This allows you to remember information about papers, create notes, and recall them later.
When to use: "remember this", "save this", "note that", "store", "don't forget", "what do I know about", "recall", "find notes about", "retrieve", "create collection", "build corpus", "gather papers", "classify", "tag", "synthesize", "compare"
This skill also provides a schema-validated, generic graph CRUD engine so agents mutate
the knowledge graph without authoring raw TypeQL. Describe writes as declarative,
schema-checked ops — typed entities, attributes, and relations — and the engine compiles and
runs the TypeQL in one atomic transaction. Commands: describe-schema, create-entity,
set-attr, delete-attr, link, unlink, delete-entity, query (guarded read-only),
apply (atomic op-list), apply-recipe (named op-list template). See USAGE.md → Generic
Graph CRUD Engine for the op vocabulary, bindings, and recipe format.
When to use: "create entity", "add a node", "set attribute", "update field", "link", "connect", "relate", "unlink", "delete entity", "run recipe", "apply ops", "describe the schema for", "what attributes does X have"
make db-startuv sync --all-extras (from project root)TYPEDB_HOST: TypeDB server (default: localhost)TYPEDB_PORT: TypeDB port (default: 1729)TYPEDB_DATABASE: Database name (default: alhazen_notebook)# Remember something
uv run python .claude/skills/typedb-notebook/typedb_notebook.py insert-note \
--subject "paper-xyz789" \
--content "Key finding: 95% editing efficiency in liver cells."
# Recall it later
uv run python .claude/skills/typedb-notebook/typedb_notebook.py query-notes --subject "paper-xyz789"
uv run emits a VIRTUAL_ENV warning to stderr. Always use 2>/dev/null when piping output to a JSON parser — never 2>&1, which merges the warning into stdout and breaks JSON parsing.
Before executing any commands, read USAGE.md in this directory for the complete command reference, workflows, and data model.