Built for Obsidian
Claude Code Plugin

Bedrock

Turn any Obsidian vault into a structured Second Brain.
7 entity types. 6 skills. Zero config.

Requires Obsidian and Claude Code

claude
>/plugin marketplace addiurykrieger/claude-bedrock
>/plugin installbedrock@claude-bedrock
>/bedrock:setup

How It Works

Teach, preserve, compress, ask

Four steps to turn scattered knowledge into a living, queryable graph.

01

Teach it your sources

Point Bedrock at a Confluence page, a GitHub repo, a Google Doc, or a CSV. It reads the source, extracts entities — people, systems, teams, topics — and classifies them automatically.

/bedrock:teach
02

It preserves the knowledge

Every entity is written to your vault with structured YAML frontmatter, hierarchical tags, and bidirectional wikilinks. One single write point — no conflicts, no duplicates, no orphans.

/bedrock:preserve
03

Compress and maintain

Over time, knowledge drifts. Compress scans your vault for duplicates, broken links, stale content, and entity misalignment — then heals it. Your Second Brain stays clean.

/bedrock:compress
04

Ask anything, anytime

Once your Second Brain is established, query it from any Claude session. Ask about systems, people, decisions, or dependencies — Bedrock searches the graph, cross-references entities, and answers with linked context.

/bedrock:ask

Commands

Six skills, one toolkit

Every skill is a specialized agent. Set up your vault, ingest sources, query knowledge, and keep everything in sync.

/bedrock:setup

Initialize vault structure and templates

📚
/bedrock:teach

Ingest from Confluence, GDocs, GitHub

🔗
/bedrock:preserve

Create entities with bidirectional links

🔍
/bedrock:ask

Search and cross-reference your vault

/bedrock:compress

Deduplicate and consolidate entities

🔄
/bedrock:sync

Re-sync with external sources

/bedrock:setup

Interactive vault initialization and configuration. Guides you through language selection, dependency checks, vault objective, and scaffolds directories, templates, config, and connected example entities.

Invoke with

/bedrock:setup

Built For

Your knowledge, structured

Bedrock adapts to any team or individual workflow where knowledge sprawl is a problem.

💻

Engineering Team Wiki

Track systems, APIs, services, and team knowledge. Every repo becomes an entity with owners, dependencies, and status — automatically linked.

📊

Product Management

Capture decisions, roadmaps, and cross-team context. Bridge notes connect people to projects to the discussions that shaped them.

🌐

Open Source Project

Document contributors, architecture decisions, and discussions. Ingest from GitHub PRs and issues directly into structured entities.

🧠

Personal Second Brain

Fleeting ideas get captured, promoted to permanent notes, and linked into your growing knowledge graph. Zettelkasten principles, automated.

The Output

Your vault, visualized

See how Bedrock organizes your knowledge into a structured, navigable graph in Obsidian.

Vault
actors/billing-api.md
---
type: actor
name: Billing API
status: active
domain: payments
pci: true
team: ["[[squad-payments]]"]
updated_at: 2026-04-15
updated_by: bedrock@agent
---
# Billing API
[!danger] PCI Scope
This actor processes card data and is subject to PCI-DSS compliance.
Core payment processing service responsible for invoice generation,
payment capture, and settlement. Integrates with [[auth-gateway]]
for token validation and [[notification-service]] for payment
receipts.
## Dependencies
-- [[auth-gateway]] — token validation
-- [[notification-service]] — email receipts
-- [[squad-payments]] — owning team
Navigating entities
Knowledge graph

Get Started

Up and running in seconds

Three commands. That's it.

claude
>/plugin marketplace addiurykrieger/claude-bedrock
>/plugin installbedrock@claude-bedrock
>/bedrock:setup
Read the docs