by affaan-mef648e01899ba3e8dc6371642deaaf64b4477775No license275K starsListed Oct 9, 2026Updated Oct 9, 2026Repository updated 4 days ago
使用此技能可防止在生产系统上工作或自主运行代理时进行破坏性操作。
Blocks destructive commands and confines file edits to chosen directories for agents working on production or autonomously.
三种保护模式：
在执行破坏性命令前进行拦截并发出警告：
检测到时：显示命令功能、请求确认、建议更安全的替代方案。
将文件编辑锁定到特定目录树：
任何在 src/components/ 之外的写入/编辑操作都会被阻止并附带说明。适用于希望代理专注于某个区域而不触及无关代码的场景。
双重保护同时生效。为自主代理提供最高安全性。
代理可读取任何内容，但仅能写入 src/api/。破坏性命令在所有位置均被阻止。
通过 PreToolUse 钩子拦截 Bash、Write、Edit 和 MultiEdit 工具调用。在执行前根据活动规则检查命令/路径。
codex -a never 会话中启用~/.claude/safety-guard.logSource:affaan-m/ECCindocs/zh-CN/skills/safety-guardat commitef648e0
License: No license
Content belongs to its original authors. SourceWeft indexes it from a public repository.
affaan-m
Designs a unified notification policy that classifies, routes, deduplicates and escalates alerts across GitHub, Linear, hooks and chat.
affaan-m
Runs a six-phase code verification loop and produces a PASS/FAIL readiness report.
affaan-m
Converts batches of UI design screenshots into Vue 3 components using Vant, Element Plus, or Ant Design Vue.
affaan-m
Guides agents in sharing durable context and handoffs between harnesses through the local ECC Memory Vault.
affaan-m
OCR-translates visa document images into English and produces a bilingual PDF with the original image and translation.
affaan-m
Interactive picker that discovers agent personas, lets a user select up to five, runs them in parallel on one task, and synthesizes the results.
anthropics
Applies a firm's KYC/AML rules grid to a parsed onboarding record, scoring risk and routing outcomes.
anthropics
Applies a firm's KYC/AML rules grid to a parsed onboarding record, scoring risk and routing outcomes.
anthropics
Tracks compliance requirements, audit readiness, and evidence for frameworks such as SOC 2, ISO 27001, GDPR, HIPAA, and PCI DSS.
Guides implementation of OAuth 2.0 DPoP (RFC 9449) sender-constrained refresh tokens for Google's OAuth platform.
Guides SOC analysts through triaging Google SecOps security alerts, from investigation to closure or escalation.
Guides SOC analysts through deep security incident and entity investigations in Google SecOps using UDM queries and timelines.
已监控的模式：- rm -rf（特别是 /、~ 或项目根目录）- git push --force- git reset --hard- git checkout .（丢弃所有更改）- DROP TABLE / DROP DATABASE- docker system prune- kubectl delete- chmod 777- sudo rm- npm publish（意外发布）- 任何带有 --no-verify 的命令/safety-guard freeze src/components//safety-guard guard --dir src/api/ --allow-read-all/safety-guard off