github bytedance/deer-flow v2.0.0
DeerFlow 2.0.0 released

4 hours ago

DeerFlow 2.0.0

DeerFlow 2.0 is a ground-up rewrite around a "super agent" harness that
orchestrates sub-agents, persistent memory, sandboxed execution,
and an extensible skills/tools system. It shares no code with the 1.x line,
which lives on the main-1.x branch.

This release closes the 2.0.0 milestone
with 182 merged PRs since the first 2.0 milestone tag.

πŸ“– Full notes: CHANGELOG.md Β· δΈ­ζ–‡η‰ˆ


⚠️ Breaking change

  • Run hydration β€” runs now hydrate from RunStore and persist interrupted
    status. Cancellation now requires the worker who owns the run; cross-worker
    cancels return 409 instead of silently appearing successful. ([#2932])

✨ Highlights

  • Custom agents that update themselves β€” agents can persist edits to their
    own SOUL.md / config.yaml from a normal chat, with full per-user
    isolation. ([#2713])
  • User-owned IM channel connections β€” users can bind their own Slack,
    Telegram, Discord, Feishu/Lark, DingTalk, WeChat, and WeCom accounts on top
    of operator-configured bots. ([#3487])
  • More models & community search β€” StepFun and MiMo reasoning models,
    MiniMax for image/video/podcast skills plus a new music-generation skill,
    and Brave Search / SearXNG / Browserless / Serper Google Images tools.
    ([#3461], [#3298], [#3437], [#3528], [#3451], [#3575])
  • Richer channels β€” Discord gets mention-only mode, thread routing, and
    typing indicators; Telegram streams replies by editing the placeholder
    message in place. ([#2842], [#3534])
  • Configurable loop detection with per-tool frequency overrides and
    deferred warning injection. ([#2586], [#2711], [#2752])
  • Subagent token usage streams to the header in real time, with spans
    attributed to the parent thread's Langfuse trace. ([#2882], [#3611])
  • Trace polish β€” clean lead_agent / custom-agent trace names; session_id
    • user_id propagated to Langfuse; non-ASCII memory traces no longer escape.
      ([#3101], [#2944], [#3104])
  • Token-usage tracking on by default, with refined display modes and usage
    attributed to the actual models. ([#2841], [#2329], [#3658])
  • Flexible memory & suggestions β€” opt out of tiktoken in restricted
    networks via memory.token_counting; make AI follow-up suggestions optional.
    ([#3465], [#3591])

πŸš€ Performance

  • Push thread metadata filters into SQL. ([#2865])
  • Index runs by thread_id in RunManager to eliminate O(n) scans. ([#3499])
  • Index messages in MemoryRunEventStore to eliminate O(n) scans. ([#3531])
  • Cache Base.to_dict column reflection per class. ([#3654])
  • Speed up should_ignore_name in glob/grep sandbox walks. ([#3657])

πŸ”’ Security

  • Reject symlinked upload destinations on Linux and Windows.
    ([#2623], [#2794])
  • Mask sensitive values in MCP config responses; harden the endpoint.
    ([#2667], [#3425])
  • Reject cross-site auth POSTs. ([#2740])
  • Cap skill artifact preview decompression (zip-bomb defense). ([#2963])
  • Mount the host Docker socket only in aio (DooD) sandbox mode. ([#3517])
  • Don't bind-mount host CLI auth dirs by default. ([#3521])

πŸ› Notable fixes

  • Memory runs on a persistent event loop instead of a short-lived one
    asyncio.run(); queued updates isolated per agent; wrapped JSON responses
    parsed correctly. ([#2627], [#2941], [#3252])
  • Runs hydrate from the persistent store after a gateway restart, return
    ISO 8601 timestamps, and have an idempotent cancel. ([#2989], [#2599], [#3058])
  • Subagents isolated from the parent run's checkpointer; timeout terminal
    state atomic; respect model overrides for tools/middleware. ([#3559],
    [#2583], [#2641])
  • Sandbox readiness polling no longer blocks the event loop;
    /mnt/user-data contract enforced at the API boundary; PVC data scoped per
    user; Windows/Git Bash MSYS path conversion disabled. ([#2822], [#2881],
    [#2973], [#2766])
  • Auth auto-generated JWT secrets persist across restarts; setup-status
    uses cached responses instead of 429. ([#2933], [#2915])
  • Frontend fixes a swallowed first message, login flicker / resize-observer
    loop, duplicate optimistic messages, deeply nested list render crashes, and
    thread isolation on new chats. ([#2731], [#2954], [#3002], [#3393], [#3570],
    [#3508])
  • Channels β€” user-owned IM now requires a bound identity, scopes files, and
    helper commands to the owner, makes the provider state authoritative, and makes
    The connect flow is deterministic, backed by shared retry helpers and
    operational guardrails. ([#3578], [#3579], [#3580], [#3581], [#3582],
    [#3583], [#3584])
  • Streaming & history β€” interrupts propagate through SSE values events for
    the LangGraph SDK, and base64 image payloads are stripped from streamed and
    history responses to keep the UI responsive. ([#3605], [#3631], [#3535])
  • Frontend (more) β€” the workspace chat list paginates beyond 50 threads,
    stays interactive when the SSR auth probe can't reach the gateway, and gets a
    mobile-friendly layout. ([#3485], [#3495], [#3646])

πŸ“¦ Deploy & ops

  • Docker Gateway defaults to a single worker (multi-worker breakage).
    ([#3475])
  • Docker nginx resolves upstream names at request time. ([#2717])
  • Packaging adds postgres extra for store/checkpointer support. ([#2584])
  • Scripts preserve uv extras across make dev restarts; clean up local
    nginx on stop; exclude runtime state from gateway reload. ([#2767], [#3005],
    [#3426])

πŸ™Œ Thanks

Huge thanks to the 40 contributors who landed 180 merged PRs in the 2.0.0 milestone, and to everyone who filed issues, tested builds, and shared feedback. DeerFlow 2.0 wouldn't exist without you.

In alphabetical order:

See the full author and PR list on the
milestone page.

Don't miss a new deer-flow release

NewReleases is sending notifications on new releases.