Bridges the agent to a persistent OKF memory vault over MCP stdio: a Python server (SQLite FTS5 + Markdown) with 10 tools for searching and storing decisions, facts, learnings, conventions, profiles and sources.
Install
# from npm (prebuilt)
dsh plugin --profile web add @luisarg/memory-mcp
# from GitHub (first run asks for allowBuilds approval — follow the hint, retry)
dsh plugin --profile web add github:Luisarg03/dsh-memory-vault#path:/packages/memory-mcp
Any plugin you install runs third-party code with your own permissions — it can read your files, use your credentials, and reach the network, and tool approvals don’t sandbox it. GitHub-sourced plugins also run build scripts at install time — pnpm blocks those until you allow them, so an install can stop with ERR_PNPM_GIT_DEP_PREPARE_NOT_ALLOWED or ERR_PNPM_IGNORED_BUILDS; dsh prints the exact key to add under allowBuilds in your profile’s pnpm-workspace.yaml, and the install works on the next run. Allowing a build is a trust decision: only install sources you trust, and pin a commit (github:owner/repo#sha).
README
Cordis bundle that connects DeepSeek Harness (DSH) to a persistent OKF memory vault over the Model Context Protocol.
It mounts @deepseek-ai/dsh-mcp-client configured for the vault's Python MCP
server (memory-vault-server in this repository) over stdio, exposing the
vault's tools (search_memory, store_*, export_memories, get_profile,
ping) to the agent.
Install
dsh plugin --profile web add @luisarg/memory-mcp
uv on PATH is recommended but not required: the bundled launcher.mjs runs the server with
uv run when uv is present and falls back to a pip-managed venv (python3 -m venv +
pip install -r requirements.txt, network on first boot) when it is not.
Skills
The package also ships the two skills that drive its tools, so installing it is enough to get the commands — nothing has to be copied into a skill directory:
| Command | Source | What it does |
|---|---|---|
/brain |
skills/brain/SKILL.md |
Reads the vault: topic search, recall by project, the profile, full exports — including the tool traps (entry_type=profile in search_memory always returns 0; tags are OR-matched; searches cap at 50) |
/checkpoint |
skills/checkpoint/SKILL.md |
Captures the session as decision/fact/learning/convention entries, then commits the vault |
They register at DSH's bundled rank (600), the weakest in the local discovery table, so a
skill of the same name in a user or project root still wins. Each SKILL.md is the single source
of its own name, description and usage guidance: the provider parses its frontmatter with yaml,
so the same file also works copied into ~/.agents/skills. Registration goes through
ctx.inject(['skills'], …), so a deployment with no skill catalog still gets the vault
bootstrap and the MCP client.
Configuration
| Variable | Purpose | Default |
|---|---|---|
DSH_MEMORY_SERVER_DIR |
directory containing server.py |
$DSH_HOME/memory-vault-server |
DSH_MEMORY_PATH |
vault directory (forwarded to the server as MEMORY_PATH) |
~/.memories |
The bundle ships no cwd-dependent paths: the cordis.patch.yml layer resolves both
values from the environment. The vault falls back to the OS home (~/.memories, which
DSH_HOME does not move); the server to the harness home ($DSH_HOME, or ~/.dsh) —
DSH does not chdir, so launching from any directory works.
Development
pnpm install
pnpm build
pnpm test
The bundle manifest (dsh.bundle) makes this package installable through the
standard dsh plugin add flow; the patch layer lives in cordis.patch.yml.
Links
More in this category
vectorize-io/hindsight#coding-agents★ 51291
Hindsight, agent memory that learns: long-term project memory with auto recall and retain, knowledge pages, deep reflection, and per-repo memory banks.
volcengine/OpenViking#examples/dsh-memory-plugin★ 40273
OpenViking memory and context bundle for DeepSeek Harness: pre-step auto-recall and profile injection, session capture, `viking://` URI guarding, and recall/write memory tools backed by an OpenViking server.
agentscope-ai/ReMe#dsh★ 3646
Connects DeepSeek Harness to ReMe's local-first, self-evolving personal knowledge base: automatically captures completed main-agent conversations as user-owned Markdown memory, searches conversations and source material through reme_search with BM25, optional embeddings, and wikilink expansion, and schedules daily memory consolidation.
zilliztech/memsearch#MemSearch★ 2759
Shared Markdown memory for DSH and other coding agents, with automatic capture, pre-step context injection, searchable recall, and memory-to-skill self-evolution through a review panel.
vshulcz/deja-vu#extensions/dsh★ 1244
Reads the session files thirty-three other coding agents on this machine already wrote — Claude Code, Codex, Cursor, VS Code Copilot Chat, opencode, OpenClaw, Hermes, Kimi, Cline, Zed and more — including sessions from before it was installed: six tools (deja_recall, deja_session, deja_blame, deja_fix, deja_how, deja_remember), a /deja command, and optional automatic recall added to the runtime context. Local BM25 index, no LLM, no embeddings, no network (dsh plugin --profile web add dsh-deja).
adoresever/graph-memory★ 655
Traceable, searchable cross-session memory for DeepSeek Harness — conversation knowledge as typed graph nodes (TASK/SKILL/EVENT) and typed edges.
Community comments
Comments are public GitHub Discussions. Loading them connects to GitHub and Giscus; a GitHub account is required to post.