TANK TASK AUTOMATION & NEURAL KNOWLEDGE

TASK AUTOMATION & NEURAL KNOWLEDGE

Built for heavy-duty work.
Driven by intelligence.

TANK is a team workspace where development agents live inside your threads. Ask for the work in a conversation; watch a plan, an approval, a pull request and a green build land in the same thread.

#platform · thread run 0x1f3a
  1. planapproved by robby
  2. branchtank/fix-nats-redelivery
  3. pr#482 opened
  4. civerify-service ✓ green

The workspace

Three panes. One objective at a time.

Navigation on the left, the Tread you are working in the middle, and a Neural Sidebar that keeps the right context and the right automations one glance away.

Tread · # platform
Goal Ship passkeys behind a flag by Friday robbymeipipeline green 4 decisions
mei WebAuthn origin check is done; needs a migration for credential ids.
tank plan Add passkey_credentials, backfill nothing, gate with auth.passkeys. Ready for approval.
robby /tank approve
/automate · /vault-pin · @tank
Neural Sidebar
Live context api/auth/passkeys.go ADR-014 · self-owned auth
Auto-Pilot Run migration check on every PR to auth/?
Instant answers Ask the Vault…

Navigation

Treads, Automation and Armor Mode. Nothing else competes for attention.

Treads

A pinned Goal Header, a filtered focus feed, threaded action cards and one-click task triggers.

Neural Sidebar

Live context from the Neural Vault, Auto-Pilot suggestions and instant answers, tied to the conversation.

Vocabulary

Stop treading water in chat rooms.

Four primitives replace the parts of chat that create noise instead of progress.

  • replaces channels

    Treads

    Objective-centric streams with a pinned Goal Header: assignees, decision log, pipeline status. You always know what a conversation is for and whether it is done.

  • replaces search

    Neural Vault

    Vector-indexed team knowledge that feeds the Neural Sidebar from the live discussion: relevant code, docs and past decisions, refreshed as the conversation moves.

  • replaces do not disturb

    Armor Mode

    Ambient chatter collapses into bulleted AI summaries per Tread. Only critical security and infra alerts break through.

  • replaces workflow builders

    Auto-Pilot

    Inline triggers such as /automate and /vault-pin turn conversation fragments into executable workflows. Conversation in, pipeline out.

Agents in the thread

Ask in the thread. Ship from the thread.

Mention @tank in any Tread. The agent plans in the open, waits at an approval gate, works in a sandbox, opens the pull request and reports CI back to the same thread. Everyone watching can steer.

  1. robby

    @tank the retry loop in worker/dispatch.go drops jobs when NATS reconnects. Fix it and add a regression test.

  2. tank agent
    Plan fix-nats-redelivery
    1. Reproduce with a reconnect test around Subscribe
    2. Ack after redelivery instead of before dispatch
    3. Add TestDispatch_Reconnect
    4. Open a PR against main
    Waiting for approval · no repo writes yet
  3. tank gate
    Approval push branch + open PR? Approved by robby · 14:02
  4. tank run
    Pull request #482 opened

    fix(worker): ack after redelivery on NATS reconnect

    +64 −12 · 2 files · sandbox run 0x1f3a
  5. tank ci
    CI green
    • lint ✓
    • test ✓
    • verify-service ✓
    Ready to merge · decision logged to the Goal Header
  1. Plan first

    The agent reads the repo and posts a plan card before it touches anything.

  2. Approval gate

    Nothing leaves the sandbox until a human approves in the thread. Edits and stops are one click.

  3. Real pull request

    Work happens in an isolated sandbox with a short-lived, least-privilege repo token. The PR is a normal PR.

  4. CI reported back

    Status flows into the same thread and the Tread's decision log, so the record of why lives with the code.

Built on

Open protocols, boring infrastructure.

No proprietary bus and no lock-in. The contracts and the SDK are public.

  • Connect RPC typed protobuf contracts, public
  • NATS JetStream events and agent intake
  • PostgreSQL partitioned by workspace
  • Temporal durable agent runs and gates
  • the agents' brain

FAQ

Questions engineers ask first.

What is a Tread?

A Tread is what TANK has instead of a channel: a stream organised around an objective. It carries a pinned Goal Header with assignees, a decision log and pipeline status, a focus feed that filters out noise by default, and action cards for the work happening inside it.

Is my code safe?

Agent runs execute in sandboxed, egress-controlled pods that are torn down after each run. Repo access uses short-lived, least-privilege tokens scoped to the run through a GitHub App. Anything that changes the world, such as pushing a branch, opening a pull request or merging, waits at an approval gate in the thread.

How do I summon an agent?

Mention @tank or use /tank in any Tread or thread. There is no ticket queue: the conversation is the brief, and the people in it steer the run.

Does TANK replace GitHub or my CI?

No. Agents work in your existing repositories and open ordinary pull requests. Your CI stays where it is; TANK reports its result back into the thread.

Which model powers the agents?

TANK agents run on frontier models through a provider-agnostic gateway, so the model can change without the product changing. The runtime is TANK's own, built on Temporal and open contracts.

Is there a mobile app?

Yes. iOS and Android apps share the same workspace, Treads and agent runs, so you can approve a plan from your phone.