Module status: Active
Infrastructure
Observability & Health Signals
Client-visible monitoring posture inside Workspace—probe status, uptime context, and health signals without exposing raw operator consoles.
Explore ModuleLoading page…
Infrastructure
Module status: ActivePolicy-gated remediation for known failure classes: your agent asks the gate before it acts, high-impact fixes wait for a human, and every decision lands on a ticket—so recurring incidents stop waiting on someone to click the same runbook.
In short: Self-Healing is the policy gate between your remediation agent and your environment. A known failure class with an enabled playbook, reachable inventory assets, and room in its blast-radius budget is allowed; Tier 2+ playbooks wait for a human to approve or veto; anything unknown fails closed to a human ticket. Every decision is audited under the agent’s name and written back to a Workspace ticket.
Every capability below maps to a live route in your client portal—not a slide.
Versioned playbooks carry trigger conditions, preconditions, rollback steps, and a blast-radius tier; organization admins enable or disable each version.
Runs bind to Workspace asset records by asset id, serial number, or asset tag—never a guessed hostname—and an unknown asset blocks the run.
Tier 2+ (medium and high impact) playbooks always wait for human approval, and a queued run can be vetoed before it fires.
Every block and every reported outcome creates a Workspace ticket with the playbook version, targets, and alert event attached.
Counted from the capabilities published for this live Workspace module.
Counted from the capabilities published for this live Workspace module.
Counted from the capabilities published for this live Workspace module.
Live in your portal
Available in your Workspace at workspace.bitscaled.tech/self-healing
Open Self-Healing in WorkspaceThe closed loop that runs for every playbook-covered failure class.
A failure class without an enabled playbook is blocked and ticketed for a human, and recurring gaps surface in catalog-gap analysis.
Every gate decision, approval, veto, and outcome is an audit event carrying the agent name and version or the deciding user.
Per-playbook asset caps, cooldowns, and run budgets are evaluated one decision at a time per playbook, so parallel alerts cannot overspend them.
Health signals and alerts are the detection input your agent classifies.
Execution targets the inventory record, not a guessed hostname.
Blocks, escalations, and outcomes are written back as Workspace tickets.
Workspace never runs shell commands. Your organization’s agent executes a playbook only after the policy gate allows it, then reports the outcome.
Every gate decision is recorded as a run with a linked ticket—there are no silent skips.
Unknown classes or assets, disabled playbooks, unmet preconditions, and exhausted budgets all block the run and open a human ticket.
Continue the Explore Module journey through adjacent ERP domains.
Infrastructure
Client-visible monitoring posture inside Workspace—probe status, uptime context, and health signals without exposing raw operator consoles.
Explore ModuleInfrastructure
Authoritative asset and inventory records in Workspace—endpoints, sites, and configuration context your team and clients can reference during incidents, onboarding, and renewals.
Explore ModuleAI Agents
Deploy governed AI agents and workflow automations from Workspace—connectors, run history, and guardrails that keep automation out of shadow IT scripts.
Explore ModuleRun the self-serve scorecard for a quick baseline, or have Bitscaled engineers review your infrastructure and security posture before you plan a rollout.
Walk through Workspace with a Bitscaled engineer—scoped to your ticket flows, asset model, and client governance requirements.