Token导航 LogoToken导航TokenDH.com
开发需要联网clawhub未标认证来源可访问clear审计提醒

llm-loop-breakerLLM loop breaker 开发

Agent Skill

llm-loop-breaker 用于补充开发相关能力,适合在 OpenClaw 中需要让 Agent 承接开发相关任务时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

2,351

周安装

97

GitHub Stars

公开资料未说明

下载量

768
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:llm-loop-breaker(LLM loop breaker 开发)
来源仓库:https://github.com/hehkcaspar/llm-loop-breaker
安装命令:
openclaw skills install llm-loop-breaker
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install llm-loop-breaker

简介

Openclaw网关的两层防御:通过熵分析杀死幻觉LLM流,看门狗主机CPU/RAM/磁盘以防止资源耗尽。

SKILL.md

name
LLM Stream Guard
description
>
metadata
openclaw
emoji
\F6E1
homepage
https://github.com/openclaw/clawhub
os
requires
bins
env
install
name
n/a
name
psutil
primaryEnv
OPENCLAW_APP_DIR
skillKey
llm-stream-guard

LLM Stream Guard

Two-layer runtime defense that protects the Openclaw gateway against hallucinating LLM streams and host resource exhaustion.

When to activate

Activate this skill once per fresh gateway deployment or whenever the gateway binary (openclaw.mjs) is rebuilt / updated.

What it does

Layer 1 -- Stream Entropy Breaker (Node.js)

Patches global.fetch to intercept every LLM streaming response (text/event-stream, application/x-ndjson, application/stream+json).

For each active stream it:

  1. Accumulates text chunks and measures compressed-vs-raw byte ratio using

zlib deflate (compression ratio = uncompressed / compressed).

  1. Every 1024 bytes, schedules an entropy check via process.nextTick.
  2. Once 4000+ bytes have been received, evaluates kill conditions:

- Hard kill: compression ratio > 10.0 - Soft kill: compression ratio > 6.0 AND single-character dominance > 50%

  1. On kill: fires AbortController.abort() with reason

HALLUCINATION_LOOP_DETECTED_BY_ENTROPY_BREAKER, severing the stream.

Bounded memory: accumulated text is truncated to the last 4096 characters when it exceeds 8192. A 30-second registry cleanup removes stale streams.

Layer 2 -- Host Resource Watchdog (Python)

Runs as a separate daemon outside the gateway process. Polls every 2 seconds via psutil.

RuleTriggerAction
Redline breachCPU >= 90% OR RAM >= 90% OR Disk >= 90%Log warning + audit entry
Poison pillChild process CPU > 95% with 0 I/O for 15 sKill process tree + incident snapshot
Memory leakRSS monotonically grows > 100 MB over 60 s, CPU > 30%Kill process tree + incident snapshot
Crash loopGateway PID changes 3+ times in 120 sIncident snapshot + 30 s cooldown

Incident snapshots capture journalctl and dmesg excerpts into ~/.openclaw/workspace/memory/core/incidents/.

Deployment

Run the deterministic deploy script. It is safe to run multiple times (idempotent).

export OPENCLAW_APP_DIR=/app          # path to openclaw installation root
bash deploy.sh

The script will:

  1. Verify python3, psutil, and $OPENCLAW_APP_DIR/openclaw.mjs exist.
  2. Copy runtime files to $OPENCLAW_APP_DIR/dist/llm_stream_guard/.
  3. Inject bootstrap code into openclaw.mjs (skipped if already present).
  4. Create a timestamped backup of openclaw.mjs before any modification.

After deployment, restart the Openclaw gateway to activate.

Verify

# Node.js layer tests (no network, no side effects)
node test-entropy-breaker.js

# Python layer tests (mocked psutil, no real process interaction)
python3 test-watchdog.py

Both test suites are deterministic and produce exit code 0 on success, 1 on failure.

Files

FilePurpose
src/stream-entropy-breaker.cjsFetch-patching + stream transform + abort logic
src/entropy-engine.cjsZlib compression ratio calculator + repetition detector
host-resource-watchdog.pySystem resource monitor daemon
deploy.shDeterministic deployment script
test-entropy-breaker.jsJS test suite (16 tests)
test-watchdog.pyPython test suite (14 tests)

Uninstall

Remove the injected block from openclaw.mjs (between the [LLM_STREAM_GUARD_START] and [LLM_STREAM_GUARD_END] markers), delete $OPENCLAW_APP_DIR/dist/llm_stream_guard/, and kill any running host-resource-watchdog.py process.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

81.44%
按下载量换算625

安全审计

VirusTotal

可疑

ClawScan

通过

Static analysis

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills