Remedy Server monitoring and remediation without unrestricted server control.
Remedy Agent is designed to run alongside customer software, collect permitted runtime evidence and execute only explicitly granted capabilities.
What Remedy Agent is designed to observe
According to configured permissions, the Agent can be designed to report application and service health, relevant logs, service status, Docker or container status, CPU, RAM and disk signals, deployed Git SHA, health-check results, API probes and other bounded runtime evidence.
Capability-based remediation
The Agent does not need unrestricted shell access. Customers determine which named capabilities Remedy receives. Examples may include restarting an approved service or container, rerunning an approved worker, executing a health verification or deploying an approved SHA where that workflow is explicitly configured.
Outbound by design
The intended model has the Agent communicate outbound to Remedy, avoiding a general inbound administration channel. Every action should be scoped, attributable and evaluated against policy.
Recovery is not always repair
A restart may restore service but leave the defect intact. Remedy is designed to verify recovery, recognise recurrence and escalate the incident into source diagnosis and governed code repair when necessary.
Frequently asked questions
Does Remedy Agent require root access?
The design is capability-based and should grant only the specific evidence and actions required, not unrestricted administrative control.
Is the Agent generally available?
The broader Agent capability is being introduced and should be treated as in development until availability is explicitly confirmed.
From bug report to verified fix.
See how Remedy connects evidence, diagnosis, repair, deterministic checks and production verification.
Explore the complete workflow →