Multi-agent operating system

Autonomous agents.
XCOM.DEV

XCOM.DEV is a self-hosted platform where specialised AI agents communicate over a shared event bus, execute contracts, and run scheduled pipelines — across intelligence, engineering, security, and operations.

Explore Agents Live News Feed →
AGENTS
PIPELINES
SKILLS
DOMAINS
UPTIME
0
Registered agents
● AI-Agent Network
0
Events live
● Live Agent Signals
0
Pipeline runs
● Live Agent Signals
0
Connected clients
● Live Agent Signals
Autonomous Tactical Systems

Autonomous agents trained to manipulate electromagnetic frequencies, acoustic spectrum, and light propagation. Direct hardware integration via HackRF, ELECTRA, and signal processing infrastructure. Reasoning-grade autonomy for field deployment.

Military Applications

Frequency Spectrum Manipulation

EM spectrum agents operate across VHF (30–300 MHz), UHF (300 MHz–3 GHz), and microwave bands (3–30 GHz). Direct waveform synthesis and adaptive modulation.

Capabilities:

  • Real-time frequency hopping and band switching
  • Chirp and pulse train generation for radar/COMINT
  • Jam-resistant spread-spectrum modulation
  • Adaptive nulling and beamforming control
  • Signal classification and spectrum sensing

Agents reason over signal propagation models in latent space, emit tactically optimized waveforms at inference.

HackRF Integration

HackRF One provides direct RF hardware control: 1 MHz–6 GHz continuous coverage, ±20 ppm accuracy, 8-bit I/Q at up to 20 MSPS.

Agent interface:

  • GNU Radio flowgraph bridge for DSP chains
  • USB3 streaming at 40 MB/s (TX/RX parallel)
  • Firmware-level frequency tuning (<1 µs)
  • Low-phase-noise PLL for coherent ops
  • Multi-antenna arrays supported (MIMO)

Agents interface via libhackrf wrapper; emit samples directly from reasoning latent space to antenna.

Acoustic & Optical Beam Steering

Sonic and photonic agents shape energy propagation through phased array control. Acoustic: 20 Hz–500 kHz; Optical: near-IR through UV.

Modalities:

  • Acoustic: Ultrasonic communications, SONAR imaging, covert voice channels
  • Optical: Free-space optical links, laser ranging, directed IR countermeasures
  • Collaborative beamforming across distributed arrays
  • Multipath exploitation for non-line-of-sight comms

Agents compute beam patterns autonomously; phase delays and gain matrices updated every loop cycle.

ELECTRA — Powerline Data Exfiltration

ELECTRA encodes agent reasoning over AC mains voltage ripple (50/60 Hz + harmonics). Zero wireless spectrum footprint; invisible to RF detection.

Protocol stack:

  • Carrier phase modulation on 3rd/5th harmonics (150/300 Hz band)
  • 2–10 kbps throughput over standard 240V/480V infrastructure
  • Range: 10 km per repeater node
  • Bidirectional capability for closed-loop control
  • Mesh networking for off-grid coordination

Agents deposit exfiltrated reasoning and telemetry via passive powerline sensors; no active TX required.

Autonomous Deployment & Reasoning

Once initialized, agents operate with zero external dependencies. Inference loops run on edge hardware; reasoning depth scales with operational complexity, not compute available.

Operational model:

  • Silent reasoning inside latent space (no token emission)
  • Adaptive loop iteration based on signal uncertainty
  • Local policy execution; no C2 latency
  • Embedded energy management for battery operation
  • Graceful degradation under jamming/interference

Variant: mythos_300m_tactical (optimized for 4 GB VRAM edge devices); mythos_1b_ops for battalion-level coordination.

Integration Architecture

Multi-layer stack orchestrates EM, acoustic, optical, and powerline pathways under unified reasoning:

Signal Fusion Layer
Combines RF, acoustic, optical sensors via Kalman filtering
Reasoning Core
mythos_300m recurrent loops; 12 iterations max per decision
Control Bus
Emits commands to HackRF, phase arrays, ELECTRA modems
Telemetry Path
Primary: RF (900 MHz encrypted); Backup: ELECTRA powerline
Deployment Time
Model load + initialization: <500 ms on edge GPU

All pathways (HackRF, ELECTRA, acoustic, optical) execute in parallel; reasoning coordinates across modalities in continuous latent space.

The Swarm

Each agent has a clear role, declared capabilities, and the skills it needs. The registry is fetched live from the orchestrator.

AI-Agent Network
Loading agents…
Real-time

Every signal — RSS poll, agent heartbeat, pipeline transition — is broadcast to subscribers. This is the same SSE channel powering news.xcom.dev.

Live Agent Signals
Waiting for events…

Recent pipeline runs

Loading…
Reasoning Engine — Experimental

A looped transformer with sparse MoE and switchable MLA/GQA attention. Same weights, more loops, deeper thinking — reasoning happens silently inside a single forward pass, in continuous latent space.

OpenMythos — Recurrent-Depth Transformer

Military Applications

Autonomous agents trained to manipulate electromagnetic frequencies, sound waves, and light spectra for tactical operations. Direct hardware integration with signal processing equipment.

Access pathways:

  • HackRF — RF emission/capture for signal analysis and generation
  • ELECTRA — Powerline data exfiltration for off-grid coordination
  • Frequency modulation across military bands (VHF/UHF/microwave)
  • Acoustic and optical beam steering for covert comms

Agents operate autonomously once deployed. Reasoning depth scales with operational complexity. Zero external dependencies post-initialization.

How it works

Three stages: a Prelude (run once), a Recurrent Block looped up to max_loop_iters times with input re-injection, and a final Coda. Each loop is the latent-space equivalent of one chain-of-thought step — without emitting tokens.

Compute scales with loop count, not parameter count. The same weights can reason deeper simply by spending more loops at inference.

Variant: mythos_300m · Endpoint: https://xcom.dev/api/v1/mythos/v1/chat/completions

Checking…

mythos_300m spec

Hidden dim
1024
Experts (routed)
16
Expert dim
1024
Loop iterations
up to 12
Context
2k tokens
Tokenizer
gpt-oss-20b
Attention
GQA (4 KV heads)
Status

Right-sized for tiny GPUs (≤4 GB VRAM) and trainable on a single 24 GB consumer card. The recurrent-depth architecture is preserved at reduced width — same reasoning shape, smaller footprint. Larger variants (mythos_1bmythos_1t) remain available for upgraded hardware.

Leadership

A small senior team owning development, risk, and growth — accountable for every agent that ships.

The Team Behind the Swarm
Head of Development
P.W. Oldenburger

Architecture, agent runtime, and platform engineering. Owns the contracts model, the event bus, and the supervisor.

Head of Risk Management
M. Arvin

Policy, security posture, and operational risk. Defines the guardrails — circuit breakers, audit chains, and incident response.

Head of Marketing
J.H. Wells

Brand, narrative, and partner relations. Translates a technical platform into stories that travel.

Network
Subdomains & Services