What problem it solves

Agent memory is often opaque: stored context cannot be easily inspected or understood. OpenViking addresses this by organizing knowledge, memory, and skills in a browsable virtual filesystem, with summaries that help agents identify relevant context before reading full files.

Key features

• Unified context filesystem — Knowledge, memory, and skills share one filesystem addressed through `viking://` URIs. Resources, memories, skills, and peers can be browsed and edited through familiar file-oriented operations.

  Source: README

• Directory-scoped retrieval — Search can be limited to a project or memory subtree instead of scanning a flat vector pool. The `find` command runs a query directly, while `search` plans retrieval from session context.

  Source: README

• Layered context loading — Directories provide L0 summaries and L1 overviews before agents need the L2 source content. This supports quick relevance checks and planning without loading every full file.

  Source: README

• Inspectable session records — Committed sessions are archived and memories are extracted as Markdown for inspection, editing, and merging. With VikingBot enabled, `ov compile` can organize source material into a wiki, knowledge graph, or report.

  Source: README

• Agent and SDK integrations — The memory plugin installer detects supported coding agents, including Claude Code, Codex, Cursor, TRAE, and OpenCode. Developers can also build integrations with the Python, Go, or TypeScript SDK, or the HTTP API.

  Source: README

Quick start

Copied verbatim from the project's README.

uv tool install openviking --upgrade && openviking-server init

ov status
ov add-resource https://github.com/volcengine/OpenViking
# Replace TASK_ID with the returned task_id; repeat until status is completed
ov task status TASK_ID
ov ls viking://resources/
ov tree viking://resources/volcengine -L 2
ov find "what is openviking"
ov grep "openviking" --uri viking://resources/volcengine/OpenViking/docs/en

pip install "openviking[bot]"
openviking-server --with-bot
ov chat   # in another terminal

Who it may suit

Inference: may suit AI-agent developers who need inspectable cross-session memory, structured knowledge retrieval, or filesystem-style context management. It may also suit teams building custom integrations through the CLI, SDKs, or HTTP API.

Topics to discuss

• How should agent knowledge, memories, and skills share a filesystem without making retrieval boundaries ambiguous?

• Which directory summary and retrieval workflows are most useful for coding agents?

• What integrations and deployment configurations should the community prioritize next?

Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/volcengine/OpenViking#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.