Dashboard
The Kyvvu dashboard is the operational interface for your compliance team. It is a vanilla JS single-page application — no framework, no build step. Serve the static files and point it at your API.
Local:
http://localhost:3000Login: use the
KV_SEED_ROOT_EMAILandKV_SEED_ROOT_PASSWORDfrom your.env
Navigation
The top navigation bar gives access to all sections:
Dashboard · Agents · Policies · Templates · Logs · Incidents · Reports · Keys
A Refresh button in the top-right reloads data from the API on demand. Data does not auto-poll.
Dashboard
The home page shows a system overview: total agents, active policies, log entries, and active incidents. A breakdown of agents by risk classification (High / Limited / Minimal / Unclassified) gives an at-a-glance compliance posture. Recent platform activity appears below.
The incidents count highlights in red when active incidents exist — the fastest way to see whether anything needs attention.
Agents
Lists all registered agents with their name, agent_id, risk classification, environment, and registration status. Click any agent to see its full metadata, recent task history, and open incidents.
Agents can be deactivated from this view. Deactivation is soft — all historical data is preserved.
Policies
Lists all policies with their name, scope, enabled status, and creation date. Filters by scope, status, and risk level.
Create Policy opens a modal with the following fields:
Policy Name and optional description
Scope —
agent_registration,step_execution, ortask_executionSeverity —
unclassified,low,medium,high, orcriticalPolicy Target — All agents, a specific agent, or all agents of a given risk classification (mutually exclusive)
Rule Type — select from available rule types; the form updates to show the relevant parameter fields
Rule Parameters (JSON) — rule-specific configuration
Enable policy — whether the policy is active immediately on save
Existing policies can be edited, enabled/disabled, or deleted. Disabling a policy stops it from evaluating against new registrations and steps — existing incidents are unaffected.
Templates
Pre-built policy collections. Each template bundles a set of related policies that can be applied in a single click — useful for quickly establishing a baseline governance framework. Applied templates create individual policies that can be reviewed and adjusted independently.
Logs
Browse execution logs across all agents and tasks. Filter by agent, environment, node type, or date range. Click a task to see its full step sequence, with input/output data and hash chain validation status shown for each step.
Each task row shows a ✓ VALID badge when the hash chain for that task is intact, along with the step count and last updated time. Expanding a task shows each step with its node_type, node_name, and timestamp.
Incidents
Lists all policy violations. Filterable by status, severity, environment, and risk level.
Each row shows the incident title, severity badge, status badge, environment, the violating agent, and when it last occurred. Clicking Resolve inline moves the incident to resolved immediately; click the incident title to open the full detail view with resolution notes and history.
The Actions sub-tab configures automated responses — create action configs (webhook, Slack, email) and map them to severity levels so incidents trigger notifications automatically.
Reports
Generate a compliance audit report for any date range. Choose format (PDF or XML) and optionally filter by environment. The report includes agent counts, task statistics, log volume, incident summary, hash chain validation results, and a configurable level of detail.
Generated reports are available for immediate download.
Keys
Manage API keys for the current account. Create new keys with a name and optional expiry. The full key value is shown only once at creation time — copy it before closing the dialog. Keys can be revoked at any time; revocation takes effect immediately and will break any agent using that key.
Last updated