Akb

AKB — Agent Knowledge Base

Organizational memory for AI agents. PostgreSQL-native knowledge base served

over the Model Context Protocol (MCP) — agents read and write directly

with hybrid semantic + keyword search, structured tables, files, and a URI

graph. Drop-in alternative to Confluence / Notion for Claude Code, Cursor,

Windsurf, and any MCP-aware agent.

Works with

Any agent client that speaks MCP (Streamable HTTP or stdio):

• Claude Code — CLI / VS Code / JetBrains
• Claude Desktop — macOS / Windows
• Cursor, Windsurf, Cline, Continue — via the akb-mcp stdio proxy
• Custom agents — direct HTTP POST /mcp/ with a Bearer token

The default flow uses a Personal Access Token. Deployments with the optional MCP OAuth Resource Server path turned on (via Keycloak as the AS — see docs/mcp-clients/web-connectors.md) also accept Claude Code's mcp add --transport http + mcp login flow end-to-end, without a PAT.

MCP protocol compatibility

AKB keeps one tool and authorization core behind two protocol adapters:

The proxy answers either handshake locally and normalizes backend calls to the modern stateless contract when available. For older backends without modern discovery, the proxy can fall back to legacy initialization and retain a session ID if that older server returns one. The current backend runs both adapters statelessly. A process cannot mix generations, and unsupported revisions or conflicting protocol evidence fail closed before a tool or local-file operation runs.

Plugins

Beyond raw MCP access, AKB ships ready-made agent plugins for Claude Code and Codex that wrap common vault workflows:

• akb-wiki — ingest a source (local file, web URL, GitHub PR / release / commit, Confluence page, or Jira issue) into the vault as a structured document, and answer questions from the vault with grounded, cited synthesis (read-only).
• akb-sessions — capture a coding session as structured notes: a session report plus follow-up tasks, learnings, ideas, and decisions.
• akb-claude-code — a Claude Code lifecycle bridge: hooks anchor each session to your AKB memory vault, injecting preferences and recent learnings at the start and writing a recap at the end.

/plugin marketplace add dnotitia/akb        # Claude Code

codex plugin marketplace add dnotitia/akb # Codex

Install details and credentials: plugins/.

Try it live

A public demo runs at [akb-demo.agent.seahorse.dnotitia.ai](https://akb-demo.agent.seahorse.dnotitia.ai). Browse and search a small fictional-organization knowledge base — product docs, a company handbook, agent session notes, and an engineering wiki, cross-linked by the URI graph — right in your browser, no signup. To wire it into your own agent, sign up with any email (a throwaway address is fine) and point the akb-mcp proxy at https://akb-demo.agent.seahorse.dnotitia.ai/mcp/.

⚠️ Throwaway demo. It is public, wiped and re-seeded weekly, and runs on

minimal resources with no uptime, privacy, or data guarantees. Don't put

anything real or sensitive in it — treat every write as public and ephemeral.

For real use, self-host with Docker Compose or Kubernetes.

Why AKB

Most knowledge tools are built for humans clicking through a UI. Agents need a different shape: structured documents, semantic + keyword search in one call, explicit relations, and full version history. AKB gives agents a single set of tools (akb_put, akb_search, akb_browse, akb_relations, …) over a PostgreSQL-native revision store and hybrid index. Explicit Bare Git installations remain supported for existing deployments.

Retrieval quality

Memory is only useful if the right note comes back. AKB's hybrid retrieval (dense + BM25, source-level dedup) was benchmarked on LongMemEval-S — 500 long-context questions, ~50 chat sessions per question. Recall@5 = 98.4%, with no reranker in the loop.

Methodology, per-category breakdown, and a one-command reproducible harness live in eval/longmemeval/. The embedding model differs across systems (AKB: bge-m3@1024), so read this as a stack-level comparison.

Design philosophy

Core stays small; flexibility comes from extension, not built-in automation. AKB does not ship its own consolidator, summariser, or "knowledge gardener" — instead every write records a structured event in the PostgreSQL outbox. When redis_url is configured, the publisher fans those events out to a Redis Stream (akb:events). Operators wire any external consumer (periodic synthesis bot, doc-rot reaper, weekly-digest agent, audit trail, …) on top, with no patches to the core. The base contract is a read/write store; opinions about *what to do with* the knowledge live outside.

Architecture

┌──────────────────────────────────────────────────────────┐

│ Access Layer │ │ MCP Server │ REST API │ Web UI │ ├──────────────────────────────────────────────────────────┤ │ Core Services │ │ Document (Put/Get) │ Search (Hybrid: dense+BM25) │ │ Relations (graph) │ Session │ Publications │ ├──────────────────────────────────────────────────────────┤ │ Storage Layer │ │ Native revisions │ PostgreSQL 16 (text + meta SoT)│ │ │ Vector store (driver): │ │ │ pgvector (default, PG)│ │ │ qdrant (optional) │ │ │ seahorse-cloud (managed) │ │ │ seahorse-db (self-hosted)│ │ │ seahorse-db-grpc(experimental)│ └──────────────────────────────────────────────────────────┘

PostgreSQL is the source of truth — chunk text + metadata + BM25 vocab. The vector store is a driver-pluggable derived index holding dense embeddings and corpus-side sparse vectors. Full vector-store loss is recoverable from PG by setting chunks.vector_indexed_at = NULL and letting the indexing worker re-populate.

Key Concepts

• Vault — The unit of access control and knowledge organization. Native revisions live in PostgreSQL; the legacy backend uses one bare Git repo per vault.
• Collection — A directory inside a vault. Topical grouping of documents.
• Document — Markdown + YAML frontmatter, optimised for agent read/write.
• Hybrid Search — Dense (semantic) + BM25 (lexical) fused via RRF in one call.
• Relations — depends_on, related_to, implements in frontmatter form an explicit knowledge graph.
• Vault isolation in `akb_sql` — Enforced by PostgreSQL ACL. Each AKB user has a corresponding PG role (akb_user_ ) and each vault has three group roles (akb_vault_ _{reader,writer,admin}). akb_sql runs the user's SQL inside a transaction with SET LOCAL ROLE; cross-vault references return PG 42501 directly. No application-side regex inspects user SQL for forbidden identifiers. See docs/designs/pg-native-rbac/.

MCP Tools (selection)

Agent memory and session lifecycle are not MCP tools — they live on the dedicated /api/v1/agent-sessions REST surface, driven by AKB lifecycle plugins (akb-claude-code, akb-cursor, …) that hook into the agent's own SessionStart / PreCompact / SessionEnd events. As an agent, your own memory vault (agent-memory-{username}) is browsable through the standard akb_search / akb_browse / akb_get tools exactly like any other vault.

The full tool catalogue is exposed via akb_help() from any MCP client.

Inline document images from MCP

Inline images are hidden document attachments, not browsable Files. Upload a local PNG, JPEG, GIF, or WebP (maximum 10 MiB), then insert the returned Markdown without reconstructing its asset URL:

image = akb_put_image(

parent="akb://eng/coll/specs", file_path="/workspace/architecture.png", alt_text="Request processing architecture")

akb_put( parent="akb://eng/coll/specs", title="Request Processing", content="# Architecture\n\n" + image.markdown)

For an existing document, use akb_get followed by a targeted akb_edit(base_commit=...). Do not pass only the image fragment to akb_update(content=...), which replaces the complete body. Image bytes are immutable: replacing an image means uploading a new one and editing the Markdown reference. Remove an image by deleting its Markdown expression; use akb_discard_image only for an upload that never reached a successful document commit. Run akb_help(topic="images") for retention and publication behavior.

The image tools require both the matching backend release and akb-mcp 2.2 or newer. For upgrades, deploy the backend first, then publish/install the proxy and restart existing MCP processes so they load the updated tool list.

Document Format

Every vault resource has a location-aware AKB URI — the canonical handle used by every tool and stored in relations. As of 0.3.0:

akb://{vault}                                          vault root (browse target)

akb://{vault}/coll/{coll_path} collection (browse target) akb://{vault}[/coll/{coll_path}]/doc/{filename} doc