dsh-task-memory

English | 中文

Task-isolated long-term memory for DeepSeek Harness.

Memories live in per-task vaults under ~/.dsh/storages/task-memory/. Facts stored for one task are invisible to another unless you deliberately switch.

Why this exists

Most DSH memory plugins are global or workspace-wide. This one treats task as the isolation boundary:

1. Default task = derived from session cwd
2. memory_bind_task rebinds the current session to a named vault
3. Search / recall / prompt injection never cross that boundary — prompt injection is registered at agent scope, so each session's system prompt only ever shows its own task's memories

Tools

Install

# From the public GitHub repo (recommended pin: full commit SHA)
dsh plugin --profile web add "github:wangyihao0001-oss/dsh-task-memory"

# Or link a local checkout while developing
dsh plugin --profile web add "$(pwd)"

After install, restart the web UI (or reboot the profile), bind a task, then remember / recall / search.

Catalog: once listed on dsh.pub, you can also install via the registry badge / dshpub flow.

Local develop (without installing)

npm install
npm run build
npm test        # node:test unit tests
npm run smoke   # build + smoke

If you run DSH from a source checkout:

pnpm dsh web --patch /absolute/path/to/dsh-task-memory/cordis.dev.yml

Update the absolute path in cordis.dev.yml so it points at this checkout’s built lib/index.js.

Config

cordis.patch.yml defaults:

injectLimit: 8           # max memories in prompt context
injectMaxChars: 2400     # soft char budget for the injected block
injectMaxEntryChars: 400 # per-entry char cap in the injected block (truncated)
injectPrompt: true       # inject pinned/recent facts for the active task
maxEntries: 500          # vault cap (>= 1); oldest non-pinned entries are evicted first (pinned are never evicted; new keys over the cap are rejected, upserts of existing keys are not affected)

Optional storageRoot overrides ~/.dsh/storages/task-memory.

Storage & reliability

~/.dsh/storages/task-memory/
  <task-id>.json

Each file:

{
  "taskId": "<task-id>",
  "title": "<title>",
  "updatedAt": 0,
  "entries": [
    {
      "id": "m_…",
      "key": "<key>",
      "content": "…",
      "tags": [],
      "pinned": true,
      "createdAt": 0,
      "updatedAt": 0
    }
  ]
}

• Files are plain JSON — safe to hand-edit or back up
• Writes go through tmp file + atomic rename, so readers always see a consistent snapshot
• Mutations for the same task (including memory_bind_task title updates, save, and update) are serialized in-process (per-task lock); concurrent agents cannot lose updates. Prefer update over load → mutate → save for read-modify-write.
• Pinned entries are never evicted; when a full vault has nothing removable but pinned entries, new keys are rejected with a clear error instead of silently dropping the just-written fact, while upserts of existing keys are never blocked by capacity (they still shrink best-effort)
• On startup, stale *.tmp files from crashed writes are cleaned up (only those older than 1h, so another process's live write is never touched)

Roadmap

• ✅ Per-session prompt injection via agent-scoped context (replaces process-level binding guess)
• Optional vector search behind the same tools
• Tiny Web UI page to browse / pin / delete vaults

License

MIT