mcpserver.lol
registry/agentwatch
Connection check local install

agentwatch

Multi-agent observability: cascade failure detection, heartbeats, and forensic replay

Tools 13
GitHub stars
Installs / wk 64
Licence MIT
Transport stdio
Last checked never

Tools & capabilities

13 tools

Read from the published package source — this server runs locally, so there is no endpoint to query. Names are taken from the code, not observed at runtime, and descriptions are often absent.

agentwatch_alert from source
Record an alert with automatic deduplication and severity escalation
agentwatch_cascade from source
Find the root cause chain for a failure event - walks parent links to find where the cascade started
agentwatch_create_trace_id from source
Generate a new trace UUID for grouping related events
agentwatch_dashboard from source
Fleet health overview - shows all agents, their status, uptime, and active alert counts
agentwatch_dashboard_text from source
Fleet health overview as a formatted ASCII table
agentwatch_fleet_health from source
Get health status of all agents with uptime percentages
agentwatch_get_alerts from source
List active (unresolved) alerts, optionally filtered by agent
agentwatch_get_failures from source
List recent trace failures, optionally filtered by agent
agentwatch_get_trace from source
Get all events in a trace by trace ID
agentwatch_replay from source
Replay all cascade chains in a trace - finds every failure and correlates each
agentwatch_report_heartbeat from source
Report a heartbeat for an agent. Call this periodically to track agent health.
agentwatch_resolve_alert from source
Mark an alert as resolved
agentwatch_trace from source
Record a trace event for cross-agent action tracking