# 📜 OpenLore

[![Release](https://img.shields.io/github/v/release/aakarim/go-openlore)](https://github.com/aakarim/go-openlore/releases/latest)
[![Go Reference](https://pkg.go.dev/badge/github.com/aakarim/go-openlore.svg)](https://pkg.go.dev/github.com/aakarim/go-openlore)

Sponsored by <a href="https://oiya.ai/?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=openlore&amp;utm_content=sponsor_logo"><img src="assets/oiya-logo.svg" alt="Oiya" height="24" align="absmiddle"></a>

**Serve your docs to AI agents over SSH.**

OpenLore is a minimal, extensible, agent-native knowledge base that keeps shared context current and inspectable.

---

## About

AI agents can already read Markdown. The problem starts when multiple agents, repositories or people need to rely on the same knowledge.

Keeping docs inside each repo works until that knowledge gets copied, duplicated or goes stale. Different agents end up working from different versions of the truth, and there is no consistent way to control who can read, update or publish what.

OpenLore gives your agents one shared place for documentation, runbooks, skills and project knowledge. Connect every agent to the same source, update it once, and make the latest version immediately available wherever it is needed.

Your knowledge stays as ordinary Markdown. OpenLore serves it as an agent-native virtual filesystem with identity-scoped access, controlled writes, validation and human approval when you need them. There is no ingestion pipeline, vector database or LLM required.

Agents can access the same knowledge through MCP or use familiar commands such as `ls`, `cat`, `grep` and `find` over SSH.

SSH is simply one interface. The important part is that every agent is working from the same current, inspectable and governed knowledge.

### Why not just keep Markdown in your repo?

For one agent working in one repository, that's fine.

OpenLore becomes useful when knowledge needs to be shared across agents, repositories or teams, or when you need permissions, publishing, review and a single source of truth without copying the same files everywhere.

[![OpenLore Skills import demo](assets/demo/v0.4.0/openlore-skills-import.gif)](assets/demo/v0.4.0/openlore-skills-import.mp4)

## Quick Start

The fastest path is to let your agent set up OpenLore:

```bash
# Teach your agent how to install, configure, and bundle OpenLore
ssh openlore.sh teach | your-agent-cli

# Add documentation access instructions to AGENTS.md
ssh openlore.sh agents >> AGENTS.md
```

Or install and run it directly:

```bash
go install github.com/aakarim/go-openlore/cmd/openlore@latest

openlore ./docs

ssh -p 2222 localhost
ssh -p 2222 localhost "grep -r 'authentication' /docs"
```

By default this starts:

- SSH on `localhost:2222`
- the human-facing web view on `http://localhost:8080`
- MCP over HTTP on `http://localhost:8080/mcp`

See [Installation](#installation) for more ways to install and package OpenLore.

## Features

- **Agent-native retrieval** — Agents use the shell tools and composition
  patterns they already understand instead of learning a bespoke retrieval API.
- **One knowledge surface, multiple transports** — Serve the same virtual
  filesystem over SSH, SFTP/SSHFS (including direct VS Code browsing and
  editing), MCP, and a human-friendly web view.
- **Live, governed knowledge** — Keep content read-only, allow scoped publishing,
  or enable full writes per docset. Writes are atomic, conflict-aware, and can
  require human approval.
- **Identity-scoped views** — Give each person or agent only the docsets it
  needs, with role-based `ro`, `publish`, and `rw` grants, path aliases, and
  private home directories.
- **Safe by construction** — The shell is an in-memory Go interpreter, not a
  real operating-system shell. There is no shell escape, arbitrary process
  execution, or ambient network access in a normal session.
- **Portable knowledge bundles** — Embed docs into a self-contained binary,
  build cross-platform bundles with the GitHub Action, or package them as a
  desktop MCP extension.
- **Structured knowledge without a new query language** — Inspect frontmatter
  as NDJSON with `lore meta`, query it with `jq`, and validate Google's
  [Open Knowledge Format (OKF)](https://github.com/GoogleCloudPlatform/knowledge-catalog/blob/main/okf/SPEC.md)
  bundles and Agent Skills close to the write path.
- **Extensible policy and processing** — Plugins can add validation, grants,
  read/write middleware, metadata, and post-commit processing while preserving
  the same filesystem interface.

## Use Cases

- **Continuous Learning repository** - store sessions and learnings in one shared server. Add metrics so you can optimise. Allow agents to share learnings with each other while maintaining user isolation.
- **Team Artifact Repository** - share markdown, HTML, JSON, Excel etc. documents you've created while maintaining access controls. Much more natural than git, more agent-native than Confluence/Notion. 
- **Documentation for coding agents** — Put internal API docs, runbooks, product
  context, and architecture notes behind a familiar, greppable interface.
- **A shared live memory for teams of agents** — Give agents separate or shared
  docsets so they can publish findings, hand off work, and accumulate durable
  context across sessions.
- **Public docs site** - add any files to your public docset, enable public access and it will be shown to any agent that stumbles across your site. Improves AEO/GEO with no need to edit your existing docs. 
- **Skills sharing** — Publish Agent Skills into shared collections so every
  authorized agent can discover and use the same governed procedures.
- **Agent Plugins repository** — Version-pin [Agent Plugins](https://agent-plugins.org) repos from GitHub and serve them to your team's agents. Skills packaged in the open standard stay current automatically.
- **Governed knowledge contribution** — Let contributors publish into inboxes
  while reserving sensitive paths for approvers and preventing accidental
  overwrites.
- **Remote review of agent artifacts** — Expose reports, logs, screenshots, and
  generated files through the browser or SSH without building a custom artifact
  viewer or granting access to the agent's machine.
- **Identity-specific workspaces** — Mount a private home for each agent plus
  shared team knowledge, all through one server and one authorization model.
- **Portable customer or project knowledge** — Ship a versioned executable with
  the relevant docs embedded, or distribute the same knowledge as an MCPB
  desktop extension.
- **Validated knowledge catalogs** — Enforce frontmatter and bundle conventions,
  inspect metadata cheaply, and stop malformed knowledge at admission time.

## How It Works

OpenLore is built on [Wish](https://github.com/charmbracelet/wish) for SSH
transport. A connection is handled entirely against a virtual filesystem:

1. **Authenticate** — connect keylessly or resolve an SSH key, certificate,
   passkey, or OAuth login to an identity.
2. **Compose a view** — mount only the docsets and paths granted to that identity.
3. **Explore** — run shell commands implemented as pure Go functions over that
   view, or use the equivalent MCP `shell` tool.
4. **Contribute safely** — if writing is enabled, authorize and validate a
   whole-file change before committing it atomically or routing it for approval.

OAuth clients use delegated identities, so durable write provenance distinguishes
direct work by `adil` from work performed as `adil/claude@claude.ai`. Delegates
can inherit no more authority than their principal and can be narrowed by
docset and capability deny lists. CIMD clients can additionally authenticate
with vendor-hosted metadata and `private_key_jwt`; see
[Authenticated OAuth Clients](docs/authenticated-oauth-clients.md).

The normal shell cannot invoke `bash`, `exec`, `curl`, or arbitrary host
processes. Embedded documentation is always read-only. Explicitly trusted
identities can be granted narrowly scoped asynchronous processing through the
`spawn` capability.

## Governed Writing

OpenLore is read-only by default. Writable deployments keep a single,
policy-controlled write path for redirects, append, `tee`, `patch`, `sed -i`,
file moves, publishing, and approved external jobs.

```bash
echo "# Research" | publish backend findings.md
cat change.diff | patch /backend/api.md
sed -i 's/old/new/g' /backend/runbook.md
```

Writes are whole-object atomic swaps. Compare-and-swap protection rejects stale
edits by default, docset grants constrain the target, and selected paths can
produce reviewable changesets under `/requests` instead of committing directly.

See [Writing and publishing](docs/writing.md) for user-facing setup and
[Write system internals](docs/write-system.md) for the implementation model.

## Installation

### Install with Go

Requires Go 1.26 or later:

```bash
go install github.com/aakarim/go-openlore/cmd/openlore@latest
```

### Build from source

```bash
git clone https://github.com/aakarim/go-openlore.git
cd go-openlore
go build -o openlore ./cmd/openlore
```

### Embed docs in a binary

Place documentation in `assets/lore/` and build. The resulting binary contains
the docs and serves them read-only at `/docs` when run with no directory
argument:

```bash
go build -o my-docs ./cmd/openlore
```

### Build with the GitHub Action

Produce cross-platform binaries with your docs embedded:

```yaml
- uses: aakarim/openlore@v1
  with:
    docs-dir: ./docs
    config: ./openlore.yml
```

See [Ways to use OpenLore](docs/usage.md) for direct VS Code editing, MCP stdio,
MCPB desktop packaging, SSHFS, and Go library usage.

### Create a customized deployment

Use the bundled `setup` skill to create `<team>-lore`, a small customer-owned
repository containing `openlore.yml`, a thin `Containerfile` pinned to an
official OpenLore release, and deployment artifacts. It builds a working local
server and verifies HTTP, MCP, authenticated SSH, writes, and persistence before
deployment:

```bash
ssh openlore.sh setup | amp
```

The generated repository keeps initial `lore.json` policy and SSH-visible files
under gitignored `.local/`. The first deployment initializes an empty persistent
volume from that state. Root `openlore.yml` remains the Git/IaC authority and is
deployed separately to `/var/lib/openlore/config/openlore.yml`; it is not baked
into the image. Later `lore.json` and filesystem edits on the server are
authoritative and are never overwritten by image updates.

Additional instruction commands support the full lifecycle:

- `onboarding` adds initial identities, roles, homes, and folders locally;
- `deploy` selects Fly.io, Railway, AWS, Google Cloud, Azure, DigitalOcean, or a
  custom deployment and verifies a shared persistence/networking contract;
- `upgrade` prepares only the pinned base-image version change so existing CD
  can deploy it.

Provider deployments require HTTPS/MCP, authenticated OpenLore SSH,
administrative shell access, and a persistent `/var/lib/openlore` volume. Where
the provider supports it, deployment configures public port 22 to forward to
OpenLore port 2222. Otherwise it reports the assigned port and recommends an
external TCP forwarding system.

The published container contains only OpenLore. It deliberately contains no
onboarding policy or server configuration. Before the service starts, the
deployment must put `openlore.yml` and `lore.json` in the persistent config
directory and run:

```bash
./out --config /var/lib/openlore/config/openlore.yml
```

This keeps configuration independently deployable: a simple deployment can
copy `openlore.yml` onto the volume, while Kubernetes can project the same file
from a ConfigMap. Use the `deploy` skill for Fly.io, Railway, AWS, Google Cloud,
Azure, DigitalOcean, or custom infrastructure. The repository's Railpack and
Fly files provide the image, persistent-volume, and port wiring; they do not
seed or mutate configuration at process startup.

Railway assigns its SSH TCP proxy a public hostname and port. Standard SSH port
22 requires an external raw TCP load balancer. Fly.io can map public port 22 to
OpenLore's internal port 2222 with a dedicated address. Raw SSH has no hostname
or SNI routing, so one listener cannot route multiple domains on port 22.

The container workflow publishes `latest` from `main`; releases also publish
`VERSION`, `vVERSION`, major, and minor image tags.

## Documentation

| Guide | Contents |
|---|---|
| [Ways to use OpenLore](docs/usage.md) | SSH, MCP, web, SSHFS, embedded binaries, GitHub Action, MCPB, and library usage |
| [Editing OpenLore files](docs/editors.md) | Direct VS Code and SFTP editor setup without a local project mirror |
| [Command reference](docs/commands.md) | Complete shell, introspection, publishing, syntax, CLI command, and flag reference |
| [Configuration and identity](docs/configuration-and-identity.md) | `openlore.yml`, authentication, roles, docsets, aliases, homes, and host verification |
| [HTTP inbox uploads](docs/inbox.md) | Upload documents with bearer or HMAC credentials |
| [Workload identity federation](docs/workload-identity-federation.md) | Authenticate CI and agents with short-lived external identity tokens |
| [Writing and publishing](docs/writing.md) | Write modes, inboxes, conflict handling, approvals, and jobs |
| [Plugins and knowledge formats](docs/plugins.md) | Plugin installation, interfaces, OKF validation, `lore validate`, and `lore meta` |
| [Folder rules](docs/folder-rules.md) | `.lore/config.yaml` and `lore.json` rules, layering, permissions, rejection messages, and growth limits |
| [Rules standard library](docs/rules-stdlib.md) | Generated reference for compiled-in rule members and their parameters |
| [Write system internals](docs/write-system.md) | Filesystem layering, write seam, changesets, hooks, and async jobs |
| [Security evaluation](SECURITY.md) | Threat model and security properties |

## Security

- Commands run in a pure-Go interpreter, not through `os/exec`.
- The virtual filesystem cleans paths and enforces docset boundaries.
- Allowed file patterns and ignored directories keep secrets out of the view.
- RBAC controls reads, publishing, writes, approvals, and trusted capabilities.
- The web endpoint can publish the SSH host key over TLS to avoid blind trust on
  first use; SSH user and host certificates are also supported.

See [SECURITY.md](SECURITY.md) for the full security evaluation.

## License

[Apache License 2.0](LICENSE) — Copyright © 2026 Adil Karim

OpenLore bundles third-party open-source components. Their licenses and required
notices are listed in
[assets/legal/THIRD_PARTY_NOTICES.md](assets/legal/THIRD_PARTY_NOTICES.md), with
full license texts in [assets/legal/licenses/](assets/legal/licenses/). These are
embedded in the binary and served by the running service at `/legal`.