HAMMABETA
GitHubDownload
Private infrastructure control

Operate every server.
Keep every command
under your control.

Direct SSH, practical server controls, and optional AI assistance in one focused app. Preview the command, understand the risk, then decide what runs.

Download HAMMA See the product
  • No account required
  • Open source
  • 5 platforms
prod-api-03 · overview● SSH SECURE

PRODUCTION / PRIMARY

prod-api-03

Healthy
CPU38%
MEMORY6.2 / 16 GB
DISK61%
ASSISTANT · LOCAL

Why is the API returning 502?

Upstream service stopped after a memory limit was reached. I found one safe recovery action.

ROOT CAUSEnode-api inactive4m ago
PROPOSED COMMANDMEDIUM RISK
sudo systemctl restart node-api

Restarts the failed service. No files or configuration will be changed.

Illustrative product flowNothing runs without approval

Works with the infrastructure you already run

LinuxDockerNginxPostgreSQLsystemdOllamaAWSHetzner

One control plane

Serious server work.
Designed for a small screen.

Use direct controls when the task is familiar. Ask the assistant when it is not. Drop into a full terminal whenever you need exact control.

••prod-api-0310.0.1.42
Connected
SERVER HEALTHAll systems operational
Updated just now
Load0.72
Memory6.2 GB
Storage61%

QUICK ACTIONS

Run health checkCPU, memory & disk
Restart a servicePreview before running
Inspect containersDocker status & logs
Review recent logsFind errors faster

Human in the loop, by design

From intent to verified result.

AI helps prepare the work. You stay responsible for execution.

  1. 01

    Connect directly

    Choose a saved server. HAMMA opens SSH from your device—there is no proxy in the middle.

  2. 02

    State the outcome

    Pick an action or explain what is wrong in plain language. The terminal remains available.

  3. 03

    Review the plan

    See every proposed command, what it changes, and a clear risk classification.

  4. 04

    Approve execution

    Run, edit, or reject. Output and the local audit trail reflect what actually happened.

INTENTVISIBLE COMMANDSYOUR APPROVALREAL OUTPUT

Built for the whole operation

Less command recall.
More server control.

HAMMA combines the practical tools used in day-to-day server management without hiding the underlying system.

01 / CONTROL

Terminal when you need it.
Clear controls when you do not.

Start with the safest interface for the job, then move into the shell without changing context.

SSH terminalVisual SFTPDocker controlssystemd unitsProcess viewerPort forwarding
02 / RUNBOOKS

Turn good fixes into repeatable operations.

Save multi-step command sequences, review each stage, and reuse the workflow when the same problem returns.

▶deploy-backend4 steps▶renew-certificate3 steps▶backup-database5 steps
03 / AI ASSISTANCE

Choose local inference or your provider.

Use Ollama, LM Studio, llama.cpp, or an explicitly configured hosted model. AI is modular; SSH is not dependent on it.

LOCAL MODEL READY 127.0.0.1
04 / EVERYWHERE

One focused experience across five platforms.

LinuxmacOSWindowsAndroidiOS

Local-first architecture

Your server credentials are not a SaaS data source.

Direct SSH remains between your device and your server. Credentials are stored in the encrypted local vault. When Local AI is selected, model traffic is restricted to loopback addresses.

  • 01
    Encrypted at rest

    AES-256-GCM vault with Argon2id key derivation.

  • 02
    Direct in transit

    SSH goes from the app to your host, not through HAMMA infrastructure.

  • 03
    Explicit at execution

    AI-generated commands are shown and confirmed before they run.

Inspect the open-source implementation ↗
DATA PATHLOCAL MODE
YOUR DEVICE
H
HAMMA APPControl plane
ACTIVE
LOCAL IPC
◆
ENCRYPTED VAULTCredentials
∿
LOCAL AI127.0.0.1
DIRECT SSH · ENCRYPTEDPort 22
▤
YOUR SERVERprod-api-03
CONNECTED

Cloud AI is optional and separately configured. It is never the SSH transport layer.

A different system boundary

Remote control without
remote custody.

HAMMA is built around device-to-server SSH. That architectural choice changes what the product needs to know about your infrastructure.

ArchitectureHAMMAManaged SSH SaaS
Connection✓Direct from your deviceThrough vendor infrastructure
Credentials✓Encrypted local vaultOften stored or brokered remotely
AI mode✓Local or explicit providerHosted model by default
Execution✓Preview, risk check, approvalVaries by product
Core SSH offline✓AvailableService-dependent

Comparison describes common managed architectures; individual vendor behavior varies by product and configuration.

Designed for responsible operators

Built for the person
who gets the alert.

01

SOLO OPERATORS

A second pair of eyes for the servers you manage alone.

Move from symptom to a reviewed command without opening six tools or pasting sensitive logs into a random chat.

02

SMALL TEAMS

Standard operations, without building a control plane.

Save runbooks, share a clear workflow, and keep raw terminal access available for the people who need it.

03

RESTRICTED NETWORKS

Useful even when cloud AI is not an option.

Use direct SSH and local inference in environments where production data cannot be sent to a hosted model.

Straight answers

Before you connect
your first server.

HAMMA is a beta release candidate and is free to inspect, download, and use.

Does HAMMA proxy my SSH connection?+

No. The core connection is made directly from the Flutter app to the server using SSH.

Can I use HAMMA without AI?+

Yes. Server profiles, SSH, terminal access, quick actions, and file operations do not require an AI provider.

Will AI-generated commands run automatically?+

No. Commands are visible before execution and require confirmation. Risky operations receive stronger warnings.

Which platforms are supported?+

HAMMA targets Linux, Windows, macOS, Android, and iOS from the same Flutter codebase.