帮助开发与运维团队快速分诊安全通告、草稿和 GHSA 报告并给出可信依据。
复制安装指令,让 AI 自动完成配置 · 推荐新手
请帮我安装 askskill 上的 "security-triage" 技能: 1. 下载 https://raw.githubusercontent.com/openclaw/openclaw/main/.agents/skills/security-triage/SKILL.md 2. 保存为 ~/.claude/skills/security-triage/SKILL.md 3. 装好后重载技能,告诉我可以用了
请分诊这份 OpenClaw 安全通告草稿,判断影响范围、是否已发布修复,并提供 shipped-tag 与 trust-model proof 作为依据,最后给出严重级别和处理建议。
一份结构化分诊结果,包含影响评估、发布状态证据、可信性说明、严重级别与建议动作。
请审查这份 GHSA 报告,核实漏洞描述是否准确,确认受影响版本与已修复版本,并用 shipped-tag 和 trust-model proof 支撑你的结论。
一份核查报告,说明报告准确性、版本影响范围、修复状态以及证据链。
请批量整理以下多个安全通告与草稿的分诊结论,按严重级别、修复状态和证据完整性输出汇总表,并标出需要人工复核的项目。
一份便于团队跟进的汇总表,突出优先级、修复进展与待复核事项。
Use when reviewing OpenClaw security advisories, drafts, or GHSA reports.
Goal: high-confidence maintainers' triage without over-closing real issues or shipping unnecessary regressions.
Close only if one of these is true:
SECURITY.mdDo not close only because main is fixed. If latest shipped tag or npm release is affected, keep it open until released or published with the right status.
Before answering:
SECURITY.md.gh api /repos/openclaw/openclaw/security-advisories/<GHSA>.git tag --sort=-creatordate | headnpm view openclaw version --userconfig "$(mktemp)"git tag --contains <fix-commit>git show <tag>:path/to/fileSECURITY.mdFor each advisory, decide:
closekeep openkeep open but narrowDefault to one advisory at a time when comments/closures are involved:
Do not batch multiple close comments unless Peter explicitly asks for a batch.
Check in this order:
SECURITY.md explicitly call this class out as out of scope or hardening-only?SECURITY.md, do not treat "injection markers" alone as a security bug.When preparing a maintainer-ready close reply:
Keep tone firm, specific, non-defensive.
…
帮助团队检索重复的 PR 与问题单,并自动分组同步重复状态。