dsh-mnemon

English · 简体中文

The three-tier, pluggable, Agent-driven memory system for DeepSeek Harness.

Three memory tiers · Nine long-term providers · One supervised workflow

Explore the capability map · Start in five minutes · Read the v0.3.2 notes · Watch the widescreen demo

dsh-mnemon gives DSH one memory control plane without forcing every kind of knowledge into one database. Runtime Memory keeps compact context available every turn. Project Documents preserve complete narratives. Memory Spaces retrieve durable evidence on demand and can use Mnemon, OpenViking, Honcho, Mem0, Hindsight, Holographic, RetainDB, ByteRover, or Supermemory.

Mnemon remains the official, prioritized native engine. The third tier is replaceable; the first two keep the same storage, workspace, and interaction model regardless of provider.

Since v0.3.0, the three tiers are the default topology of a composable kernel rather than the only shape hard-coded into every entry point. MemoryBoot wires trusted Layer, Adapter, Strategy, Guard, and MemorySource contributions into one runtime generation. Each user turn pins a lightweight TurnView: exact Runtime context enters Wake eagerly, while Documents and Memory Spaces contribute only bounded routing covers and keep complete recall authority Host-side. Users still install one dsh-mnemon package and keep the existing settings, tools, RPC, and UI workflow. See the composable architecture and extension guide.

Understand the scope in 30 seconds

The tiers are not copies. A useful rule is: every-turn context goes to Runtime, complete narratives go to Documents, and cross-task evidence goes to Memory Spaces. Current instructions, repository files, and live tool results always outrank historical memory.

Clicks that start real work

Only the rows that explicitly name a task Agent spend a separate model context; Search and Turn memory are deterministic Host reads.

Task Agents do not reuse or consume the main conversation history. By default they follow DSH's new-session model route; Settings → Memory System → Background task Agent can select a dedicated Provider and model. That fixed route also applies to bounded Mnemon workers for idle checkpoint review, writes, evidence-bound answers, provider placement, memory migration, USER compaction, Document archiving, and metadata maintenance. Conversation Recall and Related stay direct Host reads and do not use this background route.

With DSH 0.1.1-rc.2, the first-party deepseek-official/deepseek-v4-flash-vision-exp route appears with an Image input label. Mnemon background jobs still send text-only prompts; multimodal conversation messages keep their DSH attachment references through lifecycle processing, and raw image bytes are never copied into memory.

One Memory Space workflow, nine providers

Provider capability differences stay visible. dsh-mnemon never invents graph edges, deletion semantics, or enumerable content for an engine that does not provide them. Settings owns reusable Provider services; Memory Spaces owns concrete instances, activation, scope, and metadata. External Providers are off by default.

See the provider capability and deployment matrix.

Real WebUI walkthrough

The following roughly 55-second capture comes from a live 1600×900 DSH WebUI. It deliberately pauses on full-page scrolling, page transitions, Provider cards, dialogs, button-state changes, and a completed read-only Agent Query. Destructive confirmations are deliberately not submitted.

Watch the 1600×900 MP4 · Open the page-by-page UI guide

Start in five minutes

1. Install Mnemon Native

Mnemon is the default engine and the simplest local-first starting point:

# macOS
brew install --cask mnemon-dev/tap/mnemon

# macOS / Linux via Go
go install github.com/mnemon-dev/mnemon@latest

mnemon --version

Windows users can install the official v0.2.3-or-newer release ZIP. The expected installation path and checksum procedure are in Getting Started.

2. Install DSH and the plugin

This release is verified against DSH 0.1.1-rc.2, whose complete profiles require Node.js ^22.19.0 || >=24.0.0; Node 20 lacks host primitives used by rc.2. The dsh-mnemon package itself retains Node.js 20 compatibility for older compatible DSH hosts. DSH rc.2 is published on npm's latest and next tags; use the exact version below for a reproducible installation:

npm install -g @deepseek-ai/dsh@0.1.1-rc.2
dsh --version

dsh plugin --profile web add dsh-mnemon
dsh --profile web

DSH profiles have independent plugin rosters. Install the same package separately for one-shot Headless tasks:

dsh plugin --profile headless add dsh-mnemon
dsh --profile headless "Check durable project context before answering this task."

For a local checkout, use an absolute path:

dsh plugin --profile web add "link:/absolute/path/to/dsh-mnemon"
dsh plugin --profile headless add "link:/absolute/path/to/dsh-mnemon"

3. Verify the first workflow

1. Open Memory System → Status and verify dsh-mnemon, Mnemon Native, Runtime, Documents, and enabled Providers.
2. Open Memory Spaces → Overview → Create Memory Space and choose an enabled Provider explicitly.
3. Submit one stable, future-useful candidate through Remember.
4. Open Recall, run a direct search, then run Agent query against the same question.
5. Return to the conversation, expand Turn memory, and follow one exact tool link.

The primary tab order is intentionally stable: Status, Runtime, Documents, Memory Spaces.

Familiar controls, expanded capability

Agent-driven memory operations

Supervised distillation	Bounded Agent query

The workbench makes the task boundary explicit before dispatch and keeps the returned answer beside its evidence scope. Conversation-native Turn memory and Save to memory remain enabled by default and can be changed independently under Settings → Memory System → Conversation interface.

Manual or policy-driven placement

Create explicitly	Route future distillation intelligently

Manual creation always asks the user to choose. Smart selection is a distillation policy: hard rules define the eligible set, then an optional prompt guides the Agent only when several candidates remain.

Global, workspace, and custom scope

Remote Provider workspaces, users, banks, projects, containers, and URIs remain their own namespaces; switching the DSH workspace never silently rewrites them. In workspace mode, the workbench may inspect one selected workspace while the current conversation continues to execute in its own cwd. Independent task Agents launched from the workbench use the inspected workspace even when no main session is selected.

Web, conversation, and Headless share one system

Data and security boundaries

• Runtime and Documents are local deterministic stores. Mnemon Native is local by default; external Providers are explicit opt-ins.
• Provider credentials are mode 0600 under <storageRoot>/state/memory-providers.json. They are never returned to the browser, smart-selection Agent, or Mnemon Pack.
• Host calls use argument arrays with shell disabled, bounded output, timeouts, cancellation, schema validation, path boundaries, locks, and revisions.
• Disabling a Provider clears its local catalog metadata but never deletes remote data. Reconnecting rebuilds metadata from the Provider, using local defaults only when a field cannot be mapped.
• Changing scope never migrates, merges, or deletes an old root automatically.
• There is no deterministic secret scanner yet. Never store keys, tokens, private keys, or raw sensitive logs in any tier.
• Uninstalling the plugin does not remove local or remote memory data.

See Operations, security, and troubleshooting for backup, recovery, and diagnostics.

Documentation

See the documentation hub for the full map.

Development

pnpm install
pnpm run verify

verify runs TypeScript checks, Vitest, a reproducible double build, an isolated real Headless-profile activation check, and published-package validation. lib/ is generated and intentionally not tracked.

Before opening an Issue or Pull Request, read the contribution rules (简体中文) and use the bilingual repository templates. Incomplete reports and PR descriptions are rejected by bilingual automated policy checks.

License

MIT. Report security issues privately through SECURITY.md, not a public issue.