Visca

The Autonomy Cloud

Do autonomy right.

An integrated platform for building, running, and trusting autonomous systems — software agents, robots, drones, and the hybrid systems where they work together.

Verifiable

Cryptographic identity, capability grants, and audit — proof, not assertion.

Interoperable

Open protocols, framework-agnostic, embodiment-agnostic. The substrate beneath everything.

Secure by default

No agent without identity. No access without a warrant. No long-lived credential.

Architectural

One trust spine. Every actor identified, every credential ephemeral, every action recorded.

Why now

The autonomy economy is shipping. The infrastructure for it is not.

Software agents, robots, and autonomous machines are entering production simultaneously. The disciplines that govern them — identity, authorization, audit, fleet operations — were built for cloud workloads and break on the first contact with non-human, ephemeral, delegated actors.

Trust deficit

Agents hold credentials too broad to be safe.

Production agents today operate with admin tokens or service accounts with wildcard permissions. Prompt injection becomes credential exfiltration. The third option — per-action, scoped, ephemeral, identity-bound — does not exist in cloud-era infrastructure.

Operational fragmentation

Five point tools, no shared identity, no shared audit.

Agent frameworks, observability tools, sandboxes, vector stores, and tool gateways each solve a slice. None share an identity, an audit substrate, or a deployment shape. Hybrid software-plus-embodied estates run on two disjoint stacks.

No system of record

No single answer to who did what, under whose authority.

The information needed to answer accountability questions is scattered across framework traces, tool-side logs, identity events, and model-provider logs. For security and compliance, that's the blocker on production adoption.

Your agent on Visca

Same agent. Different substrate.

Pick the agent your team actually uses. See where the operational gaps are today, and how Visca closes them — without changing the agent itself.

Claude Code

Anthropic's official CLI. Top of the coding-agent leaderboards in 2026.

Claude Code site

Identity

Sigil

Claude Code runs as your shell user. Sub-agents it spawns inherit the same UID. No cryptographic root, no per-session identity, nothing to revoke.

Every Claude Code session is issued a Sigil at startup, bound to the human invoker and the runtime. Sub-agents chain their Sigils to it. One call revokes the entire chain.

Credentials

Warrant

API keys live in `.env`; MCP server tokens sit in `~/.claude/settings.json`. A prompt injection that can read the workspace reads every key.

Claude Code requests a Capability Grant per action — scoped to verb, resource, duration, and constraint. No reusable secret in any file.

Discovery

Plexus

MCP servers configured by URL in `.mcp.json`. Hardcoded endpoints, no authentication discovery, no Sigil-rooted attestation.

MCP servers found by capability descriptor through Plexus. mTLS by default. URL changes don't break anything.

5 more concerns — Declarative state, Packaging, Runtime, Audit, Dev surface.

See the full comparison
The platform

One trust spine.

Identity, access, fleet, runtime, audit, record — every operational concern of autonomy, on one substrate. Each piece has a narrow, deep scope. None overlap. All compose.

The whole is more than the sum because the pieces were never built as sums. Every product roots its security claims in the same Sigil identity and emits to the same Chronicle audit.

Visca Cloud · managed deliveryenterprise features · federation · compliance

Agent Lifecycle Management

ALM

Trust Lifecycle Management

TLM

System of Record

SoR

Open foundation · MPL 2.0

Lattice Runtime

Bundle · Sigil · Capability Grant · Plexus wire · Audit envelope. Open specifications, reference runtime, conformance suite.

Read the foundation →

ALM · Pillar

Agent Lifecycle Management

How autonomous systems are declared, packaged, deployed, and run.

TLM · Pillar

Trust Lifecycle Management

How autonomous systems are identified, authorized, and networked.

SoR · Pillar

System of Record

The queryable graph and tamper-evident audit of everything in the estate.

The business value

What organizations gain by doing autonomy right.

Visca's primitives are designed for the operational realities of autonomy: pilots that need to reach production, security teams that need accountability, finance teams that need cost governance.

3–5×
Acceleration from autonomous-system pilot to production deployment.
Projected, design-partner data
>80%
Architectural reduction in long-lived credentials held by autonomous actors.
By design — Warrant vends ephemeral, scoped grants
Single
Queryable record of every action across software and embodied actors, for security, compliance, and operations.
Chronicle, every product emits
Open foundation · MPL 2.0

Visca builds on Lattice Runtime. So can anyone.

Lattice Runtime is the open substrate beneath every Visca product. Open specifications, a reference runtime in Rust, SDKs in TypeScript, Python, Go, and Rust. Foundation-track governance. No license rug-pulls — ever.

The covenant

·

MPL 2.0 — fixed at v0.1, forever.

·

Foundation donation once the project has gravity.

·

Developer Certificate of Origin. No CLA, ever.

·

Features in the open never move behind the paywall.

·

No rent-seeking on the substrate.

The autonomy economy is shipping

Do autonomy right.

Build, run, and trust autonomous systems on an integrated platform — software, embodied, and hybrid. Open foundation. Enterprise cloud. No license rug-pulls, ever.