What makes alibabacloud-sysom-diagnosis notable?

1.4K downloads. Trust evidence available

How should alibabacloud-sysom-diagnosis be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

alibabacloud-sysom-diagnosis

Use when troubleshooting Linux server performance or stability issues — CPU saturation, high load, scheduling delay, memory pressure, OOM events, high RSS, page cache / shared memory growth, memory cgroup residue, Java heap issues, disk IO saturation or latency, packet loss, network jitter, or a server that is slow, stuck, or unstable. Performs diagnosis and surfaces recommendations; does not apply fixes automatically.

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.4k

Updated

Oct 10, 2026

Version

0.0.7

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.4K downloads reported by the source. Last updated 10/10/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 10, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 10, 2026
Adoption signal: 1.4K downloadsadoption · observed Oct 10, 2026
Latest release: 0.0.7release · observed Sep 8, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s173swjet2yrebzqrp6hjkvmy583mxef:alibabacloud-sysom-diagnosis

Install using `clawhub skill install s173swjet2yrebzqrp6hjkvmy583mxef:alibabacloud-sysom-diagnosis` in an isolated environment before connecting it to live workloads.
No published capability contract is available yet, so validate auth and request/response behavior manually.
Review the upstream CLAWHUB listing at https://clawhub.ai/sdk-team/alibabacloud-sysom-diagnosis before using production credentials.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-sdk-team-alibabacloud-sysom-diagnosis/snapshot"

Documentation

CLAWHUB

157,829 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: alibabacloud-sysom-diagnosis
description: >
Use when troubleshooting Linux server performance or stability issues —
CPU saturation, high load, scheduling delay, memory pressure, OOM events,
high RSS, page cache / shared memory growth, memory cgroup residue, Java
heap issues, disk IO saturation or latency, packet loss, network jitter,
or a server that is slow, stuck, or unstable. Performs diagnosis and
surfaces recommendations; does not apply fixes automatically.
license: Apache-2.0
compatibility: >
Requires sysom-osops CLI. The control host running the CLI can be Linux or
macOS (x86_64 or aarch64); Windows is not supported. Remote diagnosis targets
a Linux ECS instance and requires Alibaba Cloud credentials through AK/SK or
an ECS RAM Role, an online Cloud Assistant on the target ECS, and a supported
China Mainland or Hong Kong region.
metadata:
domain: aiops
product: sysom
supported_domains:
- cpu
- io
- memory
- network
- java
owner: sysom-team
contact: [email protected]
allowed-tools: Bash Read
---

# alibabacloud-sysom-diagnosis

Use SysOM CLI and backend envelopes as the diagnosis source of truth. This Skill replaces the older SysOM diagnosis Skill and is the single entry point for SysOM ECS performance and stability diagnosis.

## Immediate Route

When the user reports a symptom and has not provided fresh SysOM envelope output,
run the matching SysOM command from **Domain Routing** below before ad hoc Linux
inspection or manual probing. Then follow the returned `agent.summary`,
`agent.findings[].detail/category`, and `agent.next_steps[]`. Raw Linux commands
are bounded fallbacks only when a SysOM command is unavailable, outputs
contradict each other, or a required entity remains missing after the focused
SysOM command.

## Credential Security

Never print, echo, or ask for AccessKey ID or AccessKey Secret values. Remote
commands perform their own authentication checks. If a command returns an
authentication or permission error, explain the error and point the user to
`references/ram-policies.md`; credential setup must happen outside the
conversation.

## CLI Setup

Check whether the CLI is available:

```bash
command -v sysom-osops
```

If it is missing, install it. The installer runs on **Linux or macOS** — it does
not run on Windows. The target ECS being diagnosed must be Linux (see
`references/supported-environments.md`), but the control host can be either OS.

System-wide install (needs write access to `/usr/local/bin`, typically via
`sudo`):

```bash
curl -fsSL --connect-timeout 1000 https://sysom-prd-cn-hangzhou.oss-cn-hangzhou.aliyuncs.com/sysom_prd/skill_cli/install.sh \
| sudo bash
```

User-local install — no sudo, no root-owned paths. Works on both Linux and
macOS, and is the recommended path when you do not have administrator
privileges:

```bash
mkdir -p ~/.local/bin
curl -fsSL --connect-timeout 1000 https://sysom-prd-cn-hangzhou.oss-cn-hangzhou.aliyuncs.c

references/java/README.md

# Java Application Diagnosis Reference

Unified command for all Java diagnostics:

```
sysom-osops java analyze --type <gc|memory|cpu>
```

This document is the entry point for Java application diagnosis —
covering symptom triage, sub-domain guides, parameter quick-reference,
and combined-diagnosis strategies. Each sub-domain (gc / memory / cpu)
has its own folder under `references/java/`.

---

## Symptom Triage

### When the user's intent is unclear

If the user describes a vague Java problem (e.g., "is my Java process healthy",
"something seems wrong with my app") without mentioning specific symptoms from
the table below, ask **one** clarifying question:

> What is the primary symptom you are observing?
> 1. Application response is slow or has intermittent pauses
> 2. CPU usage is abnormally high
> 3. Memory keeps growing or OOM has occurred

Then route based on the answer:
- 1 → `--type gc` (latency-related symptoms are most often GC-induced)
- 2 → `--type cpu`
- 3 → `--type memory`

### Symptom → Diagnosis Type Mapping

When the symptom is already clear, route directly:

| Symptom | Recommended type | Description |
|---------|-----------------|-------------|
| Long GC pause / STW jitter | gc | Analyze GC event time-series and pause distribution |
| Frequent Full GC / old generation full | gc | Detect memory pressure source and promotion patterns |
| Low GC throughput / high application pause ratio | gc | Evaluate GC algorithm efficiency and tuning recommendations |
| OOM / sustained heap growth | memory | Deep diagnosis of heap/non-heap leaks |
| Native memory leak | memory | Analyze JNI/DirectBuffer usage |
| High thread CPU usage | cpu | Flame graph to locate hotspot method stacks |
| Slow application response (non-GC cause) | cpu | Analyze on-CPU time distribution |

---

## Sub-Domain Reference Index

| type | Folder / entry guide | Scope |
|------|----------------------|-------|
| gc | [gc/gc-guide.md](gc/gc-guide.md) | GC pauses, throughput, heap trend analysis |
| memory | [memory/memory-guide.md](memory/memory-guide.md) (+ `glossary`, `javamem-envelope-guide`, `decision-tree`, `profiling-playbook`, `profiling-interpretation`, `case-library` in `memory/`) | Heap/non-heap diagnosis, snapshots, allocation profiling |
| cpu | [cpu/cpu-guide.md](cpu/cpu-guide.md) | Flame graph, hotspot method stacks, on-CPU analysis |

---

## Parameter Quick-Reference

| Parameter | gc | memory | cpu |
|-----------|:---:|:------:|:---:|
| --pid | Optional | Optional | Optional (omit to get candidate list) |
| --duration | Seconds (default 60) | Minutes (0=snapshot) | N/A |
| --pod | Optional (container scenarios) | Optional | Not supported |

---

## Combined Diagnosis

When a single type cannot locate the root cause, combine multiple types:

1. **GC causing high CPU**: First `--type gc` to confirm GC frequency, then `--type cpu` to check GC thread proportion
2. **Memory leak causing frequent GC**: First `--type gc` to observe Full GC patterns, then `--t

_meta.json

{
  "ownerId": "kn74p5w8ywv6prh40g0s82gmqh83nw54",
  "slug": "alibabacloud-sysom-diagnosis",
  "version": "0.0.7",
  "publishedAt": 1788839391292
}

references/classify-output-guide.md

# Classify Output Guide

Use `sysom-osops memory classify` as the local entry point for unclear memory
symptoms. It returns the same default Agent envelope as remote deep actions.

## What To Read

| Field | Use |
|-------|-----|
| `agent.summary` | Overall memory verdict and dominant issue. |
| `agent.findings[].category` | Prefer `root_cause` when present. |
| `agent.findings[].detail` | Agent-visible root-cause entities and missing evidence. |
| `agent.next_steps[]` | Focused remote actions that can fill missing evidence. |

Do not call backend-only collectors from the Skill. Do not depend on legacy or
internal fields outside the default Agent envelope.

## Choosing The First Follow-Up

1. If a root-cause finding already contains the required entities, answer from
   classify.
2. If the dominant finding lacks a required entity, run the first relevant
   `kind=command` next step.
3. If the visible envelope names an ownership, currentness, attribution, or
   cleanup gap, follow the focused SysOM action that closes that entity gap.
4. If a later deep action returns concrete holder or owner entities, combine
   them with classify instead of re-running broad local checks.

## Next Steps May Not Be Runnable Yet

`agent.next_steps[]` can recommend remote deep commands such as
`memory filecache` or `memory memcgoffline` even though classify itself ran
locally without credentials. Those commands are registered only while the skills
catalog is reachable, so on an unconfigured machine they do not exist yet.

A `status: normal` classify result therefore does not imply the recommended
follow-up is available. If the follow-up returns domain help text or empty
output, treat it as a missing command per the rules in `SKILL.md`, tell the user
deep analysis needs credentials, and do not fall back to inventing findings or to
broad manual shell probing.

## When No Memory Action Is Needed

If classify returns no memory finding, or all findings are informational and do
not match the user's symptom, report that current SysOM memory evidence is
healthy or inconclusive. Pivot to IO, load, network, or Java only when the
visible envelope or the user's symptom supports that domain.

references/deep-actions.md

# Deep Actions Reference

This file lists public SysOM commands that this Skill may route to. On an ECS
instance, sysom-osops can auto-detect `--instance` and `--region`; add them only
for cross-instance diagnosis or when auto-detection fails.

## Memory

| Command | Mode | Use when |
|---------|------|----------|
| `sysom-osops memory classify` | Local | First route for unclear memory symptoms, OOM hints, high RSS, cache growth, shmem/tmpfs, memcg residue, or kernel memory suspicion |
| `sysom-osops memory memgraph` | Remote | Full memory landscape is missing after classify, or the issue is broad kernel/userspace memory composition |
| `sysom-osops memory memgraph --enable-socket` | Remote | Socket buffer pressure is visible and socket holder, state, PID, cgroup, or service attribution is missing |
| `sysom-osops memory process` | Remote | A process is the suspected holder but identity, real executable, cgroup, or service attribution is missing |
| `sysom-osops memory oom` | Remote | OOM killer event or memcg/host OOM evidence is visible |
| `sysom-osops memory filecache` | Remote | File/page cache is the dominant unresolved entity and file/holder attribution is missing |
| `sysom-osops memory shmem` | Remote | shmem, tmpfs, memfd, or SysV shared memory holder attribution is missing |
| `sysom-osops memory memleak` | Remote | Kernel-hidden growth (vmalloc/page-allocator/slab/percpu) is the dominant unresolved entity and the leaking call point/function/module is missing after memgraph closed only to a candidate (`--type slab\|page\|vmalloc\|percpu`, default vmalloc) |
| `sysom-osops memory javamem` | Remote | **DEPRECATED** — Use `sysom-osops java analyze --type memory` instead (see Java section below) |
| `sysom-osops memory memcgoffline` | Remote | Cgroup ownership-transition evidence is visible in SysOM output and original ownership, residue/refcount, or cleanup order needs attribution |

## IO

| Command | Mode | Use when |
|---------|------|----------|
| `sysom-osops io iofsstat` | Remote | Disk IO overview is needed for high iowait, slow disk, or IO saturation |
| `sysom-osops io iodiagnose` | Remote | Slow IO root-cause attribution is needed after the overview or when latency is the primary symptom |

## Load and CPU Scheduling

| Command | Mode | Use when |
|---------|------|----------|
| `sysom-osops load loadtask` | Remote | Load average, runqueue, or task composition is the primary symptom |
| `sysom-osops load delay` | Remote | Runnable tasks are not getting CPU time, scheduling delay is reported, or processes appear stuck without IO evidence |

## Network

| Command | Mode | Use when |
|---------|------|----------|
| `sysom-osops net packetdrop` | Remote | Packet loss, retransmits, drops, connection resets, or timeout symptoms are primary |
| `sysom-osops net netjitter` | Remote | Latency fluctuation, jitter, or intermittent connectivity degradation is primary |

## Java

Route by symptom through the unified `sysom-osops java analyze

Github ReposUpdated 21h agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.