knowledge-navigator
Read when you need to browse, traverse, or query the knowledge system beyond basic knowledge_search — graph traversal, tag exploration, orphan detection, reference browsing, topic listing, or direct SQL access.
- License 许可
- MIT
- 所属仓库
- mrtolkien/GHOST
- 最近更新
- 2026-04-24
- Stars
- ★ 5
- Forks
- 0
01 前言 Front Matter
- name
- knowledge-navigator
- description
- Read when you need to browse, traverse, or query the knowledge system beyond basic knowledge_search — graph traversal, tag exploration, orphan detection, reference browsing, topic listing, or direct SQL access.
02 资源文件 1
- 其他schema.sql6.0 KB
03 SKILL.md 正文
Knowledge Navigator Skill
This skill teaches you how to use the knowledge system's query capabilities.
Searching
Use knowledge_search for hybrid BM25 + semantic search. It defaults to notes,
references, and diary. Pass categories to narrow (e.g. ["notes"]) or extend (e.g.
add "scripts" or "code").
For CLI access: ghost knowledge search "query" [--kind note|reference|diary]
Browsing References
References are organized by topic (subdirectory name under references/).
CLI commands:
ghost knowledge references— list all references, grouped by topicghost knowledge references --topic rust— list references for a specific topicghost knowledge references --limit 50— increase result limit
Graph Traversal
The knowledge graph connects notes via typed edges ([[wiki links]]) and citations.
CLI commands:
ghost knowledge graph "Note Title"— show incoming and outgoing edgesghost knowledge graph "Note Title" --direction out— outgoing edges onlyghost knowledge graph "Note Title" --direction in— incoming edges onlyghost knowledge graph --orphans— find notes with no connectionsghost knowledge graph --stats— edge and stub counts
Tags
ghost knowledge tags— list all tags with counts
Recent Activity
ghost knowledge recent [--limit 20]— recently updated knowledge items
Stats
ghost knowledge stats— counts of notes, references, diary entries, edges, tags, and embeddings
Reference Topics
Topics are pure namespaces that group both notes and references. Notes can have an
explicit topic_id linking them to a topic namespace. Import metadata (source URL,
version) is stored in separate import_batch records, not on the topic itself.
References are organized by topic. Use the topic parameter for scoped search:
knowledge_search(query="hooks", topic="dioxus")— search all dioxus sub-topicsknowledge_search(query="pool", topic="sqlx/api")— search a specific sub-topicknowledge_search(query="framework", categories=["topics"])— search topics by name
CLI:
ghost knowledge search "query" --topic dioxus— topic-scoped searchghost topics list— list all topics with note and reference countsghost topics search "query"— search topics by nameghost reference import git --url <url> --topic <name>— import git referencesghost document import file --path <path> --topic <name>— import documents
Direct SQL Access
The schema.sql extra included with this skill contains the full table definitions for
the SQLite database. If the CLI commands above are not sufficient for debugging or
advanced queries, you can use sqlite3 directly against the workspace database:
sqlite3 ~/GHOST/ghost.db "SELECT count(*) FROM note"
Consult the schema extra to understand table structure, column names, and relationships before writing queries.
Workflow Tips
- Before creating notes, always search first to avoid duplicates.
- Use graph traversal to understand how a topic connects to existing knowledge.
- Check orphans periodically — orphan notes should be linked into the graph.
- Browse references by topic to find source material for a domain.
- Use tags to discover knowledge clusters and find related content.
04 同仓库更多 Skills
system-management
Use when reading or changing GHOST configuration (providers, models, timing, compaction, services, any config.toml key), adding or switching LLM providers, answering questions about GHOST's own features or capabilities, a CLI tool is missing or "not found", services need starting/stopping/debugging, updating GHOST to a newer version, managing the Nix shell environment, or troubleshooting GHOST setup.
writing-skills
Use when creating new skills, editing existing skills, or verifying skills work before deployment
writing-plans
Use when you have a spec or requirements for a multi-step task, before touching code. Write comprehensive implementation plans assuming the engineer has zero context for our codebase and questionable taste.
verification
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs. Requires running verification commands and confirming output before making any success claims. Evidence before assertions, always.
tdd
Use when implementing any feature or bugfix, before writing implementation code. Write the test first. Watch it fail. Write minimal code to pass.
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes