> For the complete documentation index, see [llms.txt](https://documentation.rorulabs.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://documentation.rorulabs.com/overview/roru-console.md). # Roru Console The **Roru Console** is the operator interface for Roru AI. It runs inside a local Trusted Execution Environment, meaning the interface itself — state, context, rendered output — never exists in plaintext outside the enclave boundary. Closing the console wipes session memory. Only the operator's keys can reconstitute a session. The Console is the single control plane for launching jobs, managing integrations, deploying agents, and reviewing compliance proofs. *** ## Session Model Every Console session is an isolated, attested execution environment. Before any interface element renders, the Console verifies the local enclave against the remote attestation service. A session that fails attestation does not start. **Session lifecycle:** 1. **Attestation.** The Console presents hardware attestation evidence to the attestation service. The session capsule is provisioned only after the enclave is verified. 2. **Key derivation.** Session-specific keys are derived inside the enclave. No key material crosses the TEE boundary. 3. **Context hydration.** Encrypted context from previous sessions or integrations is fetched and decrypted locally. Only hashed capability claims are transmitted — no raw data transits the network. 4. **Execution.** Operator commands compile into signed job graphs dispatched to the Executor Swarm. 5. **Session close.** On `JACKOUT` or idle timeout, enclave memory is wiped and the capsule dissolves. The Coordination Ledger retains only attestation hashes. *** ## Operator Views **Job Deck** — The primary workspace. Displays the live job graph, agent execution status, encrypted chat with AI operators, and timestamped attestation events. All state is decrypted in-memory and never written to the host filesystem. **Integrations view** — Visualizes all connected Secure Data Connectors with live policy status, rate limit telemetry, and anomaly alerts from Enclave Integrity Monitors. Each integration's policy capsule — scope, retention window, expiration — is inspectable and modifiable here. **Signal Lattice View** — A composable analytics dashboard where encrypted datasets are stitched together into operational intelligence. Rendered exclusively within the active session; the rendered view is destroyed on close. **Executor Bay** — Agent management panel. Operators can inspect active agents, review their current attestation state, adjust policy guardrails, and view per-step compliance receipts in real time. **Audit Trail** — Read-only view of zero-knowledge receipts emitted by the job. Receipts are exportable and verifiable by any party with access to the public verification key, without access to job content. *** ## Console Security Properties **In-memory rendering.** UI elements are rendered inside enclave memory. A screen capture tool running outside the TEE captures only static noise — the host cannot read the interface. **Ephemeral session state.** Session state is never written to disk. On close or timeout, the enclave is wiped. Saved job templates are stored encrypted under the operator's key — the Console cannot open them without key authorization. **Presence obfuscation.** Collaborators in a shared session are identified only by pseudonymous operator tokens until a mutual disclosure handshake occurs. External observers cannot determine who is present in a session or when they connected. **Policy-locked command execution.** Commands are compiled into deterministic job graphs before dispatch. The compiled graph is hashed and committed to the audit trail prior to any agent receiving it — the operator can inspect and approve the exact execution plan before it runs. *** ## Console Shortcuts | Command | Action | | ------------------------------ | ------------------------------------------------------------------ | | `CTRL + REDPILL` | Open the command palette to compose and dispatch a new intent | | `JACKIN` | Start a new authenticated session | | `JACKOUT` | Close session, wipe enclave memory, and dissolve the capsule | | `AGENT.SEND ` | Dispatch a reconnaissance or execution agent to a specified target | | `SENTINEL ` | Register an alert rule on an integration or agent behavior pattern | *** ## Next Steps * Follow [Getting Started](/guides/getting-started.md) to configure and launch your first job. * Connect external systems via the [Integration Guide](/guides/integration-guide.md). * Review [Platform Capabilities](/product/features.md) for a complete feature reference. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://documentation.rorulabs.com/overview/roru-console.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.