Argus CLI Command Reference
May 29, 2026 · View on GitHub
Complete reference for all 71 Argus CLI commands. Commands are organized into five categories. Each category page contains full synopses, option tables, and output samples. Use the alphabetical index at the bottom to jump directly to any command.
Global Options
--source=auto|agent|jdk Data source (default: auto)
--no-color Disable ANSI colors
--lang=en|ko|ja|zh Output language
--format=table|json Output format (default: table)
--host HOST Agent host (default: localhost)
--port PORT Agent port (default: 9202)
--help, -h Show help
--version, -v Show version
Locale resolution order: --lang=<code> flag > $LC_ALL > $LANG > ~/.argus/config.properties (set by argus init) > en fallback.
JSON Output
All commands support --format=json for scripting and pipeline integration:
$ argus --format=json gc 39113
{"totalEvents":0,"totalPauseMs":0.0,"overheadPercent":0.0,"lastCause":"","heapUsed":56849408,"heapCommitted":268435456,"collectors":[{"name":"g1 gc","count":0,"totalMs":0.0}]}
Multi-language Support
Set the output language with --lang or configure it with argus init:
$ argus --lang=ko histo 39113 --top 3
Supported languages: English (en), Korean (ko), Japanese (ja), Chinese (zh)
Categories
Monitoring
Real-time process observation, dashboards, environment inspection, multi-instance cluster health, and alerting. → commands/monitoring.md
argus ps— list running JVM processesargus info— JVM version, uptime, VM flagsargus env— launch environment (classpath, VM args)argus top— real-time agent dashboardargus watch— htop-style live JVM dashboardargus report— single-view diagnostic reportargus diff— heap snapshot comparison (leak detection)argus alert— threshold-based webhook alertingargus cluster— multi-instance aggregated healthargus perfcounter— low-level JVM performance countersargus tui— interactive terminal UI (k9s-style)argus harness— continuous trend-aware health watch (doctor + 4 trend rules over a window)
Memory & GC
Heap inspection, GC statistics, NIO buffers, metaspace, classloader leaks, GC log analysis, and ZGC health verdicts. → commands/memory-gc.md
argus gc— GC stats (heap usage, pause time, collector)argus gcutil— generation utilization % (jstat-style)argus gcnew— young-gen detail (eden, survivor, tenuring)argus gccause— last/current GC cause with gen utilizationargus gcrun— trigger System.gc() remotelyargus heap— detailed heap usage with space breakdownargus heapdump— generate heap dump (.hprof)argus histo— heap object histogram (count + size)argus nmt— native memory tracking by categoryargus buffers— NIO direct/mapped buffer poolsargus finalizer— finalizer queue statusargus metaspace— metaspace usage by space typeargus classstat— class loading throughputargus classleak— classloader-level metaspace leak attributionargus gclog— analyze GC log files (pauses, causes, tuning)argus gclogdiff— compare two GC log filesargus gcscore— A–F GC health scorecardargus gcwhy— narrate the worst GC pause in plain Englishargus gcprofile— JFR-based allocation profilingargus heapanalyze— offline HPROF analysis (MAT alternative)argus g1— G1GC live health verdict (HEALTHY/WARNING/UNHEALTHY)argus zgc— ZGC live health verdict (HEALTHY/WARNING/UNHEALTHY)
Profiling & Tracing
CPU, allocation, lock, and wall-clock profiling via async-profiler; flame graph generation; JFR control; slow method detection; and method call tracing. → commands/profiling-tracing.md
argus profile— CPU/alloc/lock/wall profiling (async-profiler)argus profile-gate— CI/CD CPU regression gateargus flame— one-shot flame graph (HTML, opens browser)argus jfr— JFR start/stop/check/dumpargus jfranalyze— analyze JFR recording fileargus slowlog— real-time slow method detection (JFR streaming)argus trace— method call tree from thread-dump samplingargus instrument— opt-in live method instrumentation (watch args/return/exception, trace call tree, monitor invocation stats)argus benchmark— sampling-based throughput estimate for a methodargus snapshot— forensic bundle (threads + histogram + doctor + JFR + optional heap dump) as a single tar.gz
Runtime & JVM Internals
VM flag inspection and live modification, unified logging control, JMX management, JIT compiler state, class search, Spring Boot integration, and cross-cutting operations tooling. → commands/runtime-internals.md
argus vmflag— show/filter non-default VM flagsargus vmset— set a manageable VM flag at runtimeargus vmlog— control JVM unified logging at runtimeargus sysprops— JVM system properties (-D flags)argus compiler— JIT compiler status + code cache usageargus compilerqueue— current JIT compilation queueargus mbean— MBean browser (domains, attributes)argus jmx— JMX management agent controlargus dynlibs— native libraries loaded in the JVMargus stringtable— interned string table statisticsargus symboltable— JVM symbol table statisticsargus sc— search loaded classes by patternargus classloader— class loader hierarchy + class countsargus spring— Spring Boot app inspection via JMXargus logger— view/change log levels at runtimeargus events— VM internal event log (safepoints, deoptimizations)argus explain— explain JVM metrics, GC causes, and flagsargus doctor— one-click JVM health diagnosisargus suggest— JVM flag optimization by workloadargus ci— CI/CD health gate with exit codesargus compare— side-by-side JVM comparisonargus init— first-time setup wizard (language preference)
Threads
Thread state summaries, full thread dumps, deadlock detection, and thread pool distribution analysis. → commands/threads.md
argus threads— thread state summary with distribution barsargus threaddump— full thread dump (jstack replacement)argus deadlock— automated deadlock detectionargus pool— thread pool grouping with state distribution