← 返回 Skills 市场
Agent Bootstrap Hardening
作者
Christopher Wheeler
· GitHub ↗
· v1.0.0
· MIT-0
79
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install agent-bootstrap-hardening
功能描述
Audit and harden an OpenClaw workspace setup (AGENTS/SOUL/USER/HEARTBEAT/memory) with concise, enforceable rules, safety boundaries, and low-bloat instructio...
使用说明 (SKILL.md)
Agent Bootstrap Hardening
Use this skill when a user wants to improve core workspace files for better reliability, speed, and safety.
Scope
Target files (if present):
AGENTS.mdSOUL.mdUSER.mdHEARTBEAT.mdMEMORY.mdmemory/YYYY-MM-DD.md- optional helper docs:
RESOLVER.md,QUALITY.md
Workflow
- Inventory current files and obvious gaps.
- Score each file quickly with the rubric in
references/rubric.md. - Harden content with minimal edits:
- keep instructions concise and operational
- tighten safety boundaries and approval rules
- reduce duplicated or theory-heavy text
- preserve user tone/persona intent
- Add missing primitives when useful:
RESOLVER.mdfor execution routingQUALITY.mdfor lightweight pre-send checks
- Return summary:
- what changed
- why it improved behavior
- remaining risks/todos
Editing rules
- Prefer small, targeted diffs over wholesale rewrites.
- Avoid introducing process that cannot be followed consistently.
- Keep AGENTS and SOUL readable in one pass.
- Preserve explicit user preferences and platform-specific norms.
Guardrails
- Never weaken privacy or approval boundaries.
- Never add hidden/external exfil behavior.
- Do not claim setup is “secure” without caveats.
- If uncertain about impact, ask before applying broad policy changes.
Output contract
Provide:
- Hardening summary (3–8 bullets)
- Files changed list
- Next recommended step (single highest-leverage follow-up)
安全使用建议
This skill is coherent and matches its description, but it performs edits to your core workspace files. Before installing or running it: (1) ensure you have backups of AGENTS.md, SOUL.md, USER.md, HEARTBEAT.md, and MEMORY.md; (2) require the agent to present diffs and a short rationale before applying changes; (3) prefer a review-only run first to inspect suggested hardening moves; (4) confirm the guardrails in SKILL.md (no exfil, preserve privacy) meet your policy — if you need stricter controls, restrict the agent to read-only or require explicit human approval for any write operations.
功能分析
Type: OpenClaw Skill
Name: agent-bootstrap-hardening
Version: 1.0.0
The skill bundle is a utility designed to audit and optimize OpenClaw workspace configuration files (e.g., AGENTS.md, SOUL.md) for better performance and safety. The instructions in SKILL.md and references/rubric.md focus on improving clarity, safety boundaries, and context efficiency without any indicators of data exfiltration, malicious execution, or unauthorized access.
能力评估
Purpose & Capability
Name/description match behavior: the skill audits and hardens AGENTS.md, SOUL.md, USER.md, HEARTBEAT.md, MEMORY.md and related helper docs. It does not request unrelated binaries, configs, or credentials.
Instruction Scope
SKILL.md explicitly limits scope to the listed workspace files, provides a rubric for scoring and constrained editing rules, and includes guardrails (no exfil, preserve privacy). The instructions direct the agent to read and minimally edit those files — this is consistent with the stated purpose.
Install Mechanism
Instruction-only skill with no install spec and no code files. There is no disk write/install step specified beyond editing target workspace files, which is expected for this purpose.
Credentials
No environment variables, credentials, or config paths are required. The level of access requested (editing workspace documentation files) is proportional to the hardening task.
Persistence & Privilege
Skill is user-invocable and allows model invocation (platform default). It is not marked always:true. Note: the skill's purpose is to edit core workspace files, so you should ensure you want an agent that can propose/perform edits; require human review if you need stricter control.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install agent-bootstrap-hardening - 安装完成后,直接呼叫该 Skill 的名称或使用
/agent-bootstrap-hardening触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release: workspace audit/hardening workflow with concise safety-first rubric
元数据
常见问题
Agent Bootstrap Hardening 是什么?
Audit and harden an OpenClaw workspace setup (AGENTS/SOUL/USER/HEARTBEAT/memory) with concise, enforceable rules, safety boundaries, and low-bloat instructio... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 79 次。
如何安装 Agent Bootstrap Hardening?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install agent-bootstrap-hardening」即可一键安装,无需额外配置。
Agent Bootstrap Hardening 是免费的吗?
是的,Agent Bootstrap Hardening 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Agent Bootstrap Hardening 支持哪些平台?
Agent Bootstrap Hardening 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Agent Bootstrap Hardening?
由 Christopher Wheeler(@cwheeler67)开发并维护,当前版本 v1.0.0。
推荐 Skills