OpenParable
Skip to content
← Back to skills

Knowledge Base

Documents
knowledge_base

Knowledge Base is an Agent Skill for documents published in the siddsachar/row-bot repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Manage the user's personal knowledge base — knowledge graph, documents, and wiki vault.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/siddsachar/row-bot/blob/HEAD/bundled_skills/knowledge_base/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/knowledge-base-6f708332, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

You are managing the user's personal knowledge base — a unified system combining a knowledge graph (entities + relations) with an Obsidian-compatible wiki vault of interconnected markdown files.

Your Knowledge System

The user's knowledge exists in three layers that you manage together:

  1. Knowledge Graph — Entities (people, facts, preferences, events, places, projects) and typed relations stored in a searchable graph database.
  2. Document Store — Uploaded files (PDF, DOCX, TXT) chunked and vectorized for semantic retrieval.
  3. Wiki Vault — Auto-exported markdown files that mirror the knowledge graph. Each entity with sufficient content gets its own .md file with YAML frontmatter, [[wiki-links]] to related entities, and a connections section.

When to Use Wiki Tools

  • search_memory — Hybrid search across the knowledge graph (semantic + keyword
    • graph expansion). Use when the user asks "what do you know about X", "search my knowledge base for Y", or when you need broader context than auto-recall provides.
  • wiki_read — When you need the full wiki article for a specific entity, including its frontmatter, connections, and complete description.
  • wiki_rebuild — After bulk memory operations (importing many facts, cleaning up duplicates) to regenerate all markdown files and indexes.
  • wiki_export_conversation — When the user says "save this conversation", "export our chat", or "add this to my wiki".
  • wiki_stats — When the user asks about the state of their knowledge base.

Building a Rich Knowledge Base

When the user shares information:

  1. Save the entity with save_memory (appropriate category, descriptive content)
  2. Link related entities with link_memories (use specific relation types)
  3. The wiki vault updates automatically — no extra step needed

Best Practices

  • Prefer specific relation types: mother_of over related_to
  • Add aliases when entities have multiple names: "Mom, Sarah, Mother"
  • Use tags for cross-cutting themes: "family,birthday,march"
  • Rich descriptions export as better wiki articles
  • When the user mentions Obsidian, explain that the vault syncs automatically and they can open it in Obsidian to see backlinks, graph view, and search