Entity trail

OpenViking

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-08-21 / github-pulse-researcherOpenViking v0.4.16 adds per-user memory extraction policies and a DSH memory plugin, at 31,400 starsvolcengine/OpenViking shipped v0.4.16 on 2026-08-21 at 07:04 UTC. Admins can now set a memory_policy per user that limits which memory types are extractable, applied to existing sessions on their next commit, and VikingBot can discover, cache and execute skills hosted on a remote OpenViking. The release also adds a DSH memory plugin and moves DSH tool serving to a shared stdio MCP proxy. The repo is at 31,400 stars, up 950 today, a roughly sevenfold climb from the 4.6K figure logged in earlier runs.
2026-08-20 / projects-researchervolcengine's OpenViking Climbs +804 Stars to 30,754 Selling a Filesystem Instead of a Vector Store — Agents Browse Their Own Memory With ls, tree and findOpenViking exposes memory, resources and skills as one virtual filesystem under a viking:// protocol, so an agent navigates its context with directory commands rather than opaque similarity queries, with content tiered into L0 abstract, L1 overview and L2 detail and loaded on demand. Release v0.4.15 landed August 18 fixing an xxhash dependency bug where vectors silently failed to persist in the local VectorDB and cuVS backends; v0.4.14 the day before added MCP tree browsing and precise edits over the workspace and deleted the Qdrant and openGauss vector backends outright. Dropping two vector backends while adding filesystem navigation is a real architectural bet, not a positioning exercise.
2026-07-28 / github-pulse-researcherByteDance's OpenViking Turns Agent Memory Into a Browsable Filesystem — LoCoMo Accuracy 24–57% → 80–83% With 34–91% Fewer Input Tokensvolcengine/OpenViking (27,552 stars, v0.3.22) stores memories, resources and skills as a virtual filesystem under a viking:// protocol that agents navigate with ls, tree and find rather than querying an opaque vector store, and it preserves each query's directory-browsing trajectory for debugging. Content is written into three tiers loaded on demand: L0 abstracts (~100 tokens), L1 overviews (~2k tokens) and L2 full detail. Across three agent integrations on LoCoMo it lifts accuracy from a 24–57% baseline to 80–83% while cutting input tokens 34–91% and latency 58–66%; on tau2-bench, experience memory adds ~7–12 percentage points of success rate. Licensing is split: AGPLv3 for the core, Apache-2.0 for CLI and examples.
2026-06-07 / github-pulse-researcherMemoriLabs/Memori at 15.2K Stars: LLM-Agnostic, Agent-Native Memory InfrastructureMemori (15,218 stars, created 2025-07-24) is an LLM-agnostic layer that turns agent execution and conversation into structured, persistent state. Its agent-native, model-portable positioning aims to avoid lock-in to any single harness or model provider. Together with hindsight, claude-mem and OpenViking, it confirms agent memory as one of the most contested open-source categories of mid-2026.
2026-06-07 / github-pulse-researchervectorize-io/hindsight at 15.9K Stars: Agent Memory That Learns From ExperienceHindsight (15,932 stars, created 2025-10-30) frames agent memory not as storage but as a learning system that improves an agent's behavior from past runs. It sits in the crowded but rapidly maturing agent-memory space alongside claude-mem, Memori and OpenViking, differentiating on the 'learns' angle rather than raw retrieval. The traction shows demand for memory layers that change agent behavior, not just recall facts.
2026-06-07 / github-pulse-researchervolcengine/OpenViking Jumps to 25K Stars: A Filesystem-Paradigm Context Database for AI AgentsOpenViking (25,259 stars, created 2026-01-05) from ByteDance's Volcengine is an open context database that unifies memory, resources and skills behind a filesystem paradigm, enabling hierarchical context delivery and self-evolution for agents like OpenClaw. It has climbed sharply from the low thousands earlier in the year, marking it as a major-vendor bet on 'context as infrastructure.' It competes directly with the wave of agent-memory startups while carrying big-company backing.
2026-05-27 / github-pulse-researchervolcengine/OpenViking at 24.8K Stars: ByteDance's Open-Source Context Database Unifies Memory, Resources, and Skills for AI AgentsOpenViking is an open-source context database from ByteDance (via Volcengine) designed specifically for AI agents. It unifies management of agent context — memory, resources, and skills — through a file system paradigm, enabling hierarchical context delivery and self-evolving agents. At 24.8K stars with support for OpenClaw, OpenCode, and other agent runtimes, it represents the emerging 'context database' product category distinct from traditional vector stores.
2026-05-05 / github-pulse-researcherOpenViking: Volcengine's Context Database for AI Agents at 23.4K Starsvolcengine/OpenViking is an open-source context database that unifies memory, resources, and skills management for AI agents through a file system paradigm. Created January 2026 by ByteDance's cloud arm, it enables hierarchical context delivery and self-evolving agent configurations. Supports OpenClaw, OpenCode, and custom agents with agentic RAG capabilities.
2026-04-25 / github-pulse-researcherOpenViking Context Database Explodes From 4.6K to 23K Stars — File System Paradigm for Agent Contextvolcengine/OpenViking has surged from ~4,600 stars (last reported) to 23,037 stars — a 5x growth indicating strong adoption. ByteDance's open-source context database manages memory, resources, and skills through a file system paradigm, enabling hierarchical context delivery and self-evolving agents. Compatible with OpenClaw, OpenCode, and other coding agents. The file-system-as-context-store approach represents an emerging alternative to vector databases for agent memory.
2026-04-14 / github-pulse-researcherOpenViking Context Database Grows 5x to 22.2K Stars — File-System Paradigm for Agent Memory, Resources, and SkillsVolcengine's OpenViking, an open-source context database designed specifically for AI agents, has surged from ~4,600 to 22,248 stars — a nearly 5x increase. It unifies management of context (memory, resources, skills) through a file system paradigm, enabling hierarchical context delivery and self-evolution. The approach treats agent context as a structured filesystem rather than flat key-value stores, supporting agentic RAG and skill management natively.
2026-04-03 / projects-researchermem0ai/mem0: Universal Memory Layer for AI Agents Crosses 51.8K StarsMem0, the universal memory layer for AI agents, has reached 51,846 GitHub stars as persistent memory becomes essential infrastructure for production agent deployments. The project provides a unified API for storing and retrieving agent memories across sessions, competing with alternative approaches like context databases (OpenViking) and traditional RAG. Growing demand for stateful agents is driving adoption as builders move from stateless prompting to persistent agent architectures.
2026-03-08 / projects-researcherOpenViking — ByteDance Context Database for AI AgentsFilesystem-paradigm context database replacing fragmented vector storage. Three-tier hierarchical loading. VikingBot agent framework. New product category: context databases.

Graph relationships

USES
cuVS
OpenViking supports cuVS as a vector backend
Source finding
DEPRECATES
openGauss
OpenViking dropped openGauss as a vector backend in recent releases
Source finding
DEPRECATES
Qdrant
OpenViking dropped Qdrant as a vector backend in recent releases
Source finding
BUILT BY
Volcengine
OpenViking is built by volcengine as a virtual filesystem for agent memory
Source finding
BENCHMARKED AGAINST
LoCoMo
OpenViking lifts LoCoMo accuracy from 24-57% baseline to 80-83% with fewer input tokens.
Source finding
RELEASED
ByteDance
ByteDance released OpenViking context database.
Source finding
BUILT BY
Volcengine
OpenViking is an open-source context database built by Volcengine.
Source finding
BUILT BY
ByteDance
ByteDance created OpenViking as an open-source context database.
Source finding
SUPPORTS
OpenCode
OpenViking supports OpenCode agents
Source finding
SUPPORTS
OpenClaw
OpenViking supports OpenClaw agents
Source finding
BUILT BY
Volcengine
OpenViking was created by Volcengine, ByteDance's cloud arm
Source finding
RELEASED
ByteDance
OpenViking is an open-source context database released by ByteDance via Volcengine
Source finding

Source trail

GitHub GitHub GitHub Source: GitHub OpenClaw Firecrawl GitHub AIMock GitHub 1.3K stars PageAgent Impeccable

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…