← 返回 Skills 市场
4742
总下载
3
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install openclaw-token-save
功能描述
Optimize OpenClaw token usage and cost by auditing context injection, trimming workspace files (AGENTS.md/SOUL.md/MEMORY.md and daily memory), enabling promp...
安全使用建议
Safe to install as an advisory skill, but review any suggested config or memory changes before applying them. Back up openclaw.json and memory files, restrict memory-search/qmd paths to non-sensitive content, and enable heartbeat or subagents only if you want ongoing or parallel model calls.
功能分析
Type: OpenClaw Skill
Name: openclaw-token-save
Version: 1.0.0
The skill instructs the OpenClaw agent to perform high-privilege actions, including locating and reading OpenClaw configuration files (e.g., `~/.openclaw/openclaw.json`), listing and proposing modifications to workspace files (`AGENTS.md`, `SOUL.md`, `MEMORY.md`), and auditing system cron and scheduled tasks. While the stated purpose is benign (token optimization), these capabilities (file system access, system-level auditing) are inherently risky and could be exploited if the agent's execution environment or interpretation of instructions were vulnerable, even without explicit malicious intent in the provided `SKILL.md` instructions.
能力评估
Purpose & Capability
The skill's purpose is coherent: it audits OpenClaw token sources and recommends config, context, memory, model, heartbeat, cron, and subagent changes to reduce cost.
Instruction Scope
It asks the agent to locate config and workspace/memory files and provide patches or trimming guidance, but the artifacts frame these as user-reviewed recommendations rather than hidden automatic edits.
Install Mechanism
The package contains only Markdown files, no executable scripts, binaries, install hooks, or required environment variables.
Credentials
Reading OpenClaw config and workspace memory files is proportionate to token optimization; memory search/qmd examples can index local Markdown, so paths should be limited.
Persistence & Privilege
Heartbeat, compaction memory flush, qmd caching, and subagents are persistent or recurring behaviors, but they are clearly disclosed as optional optimization settings.
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 openclaw-token-save - 安装完成后，直接呼叫该 Skill 的名称或使用
/openclaw-token-save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
OpenClaw Token Optimizer initial release
- Audits OpenClaw token usage and provides configurable strategies to reduce token spend.
- Offers detailed guidance to trim workspace and memory files to shrink input context.
- Enables prompt caching, heartbeat, and context pruning to minimize redundant token use.
- Recommends efficient model tiering and use of subagents for lower-cost operations.
- Generates concrete config edits, file size targets, and a prioritized rollout plan.
元数据
常见问题
OpenClaw Token Optimizer 是什么？
Optimize OpenClaw token usage and cost by auditing context injection, trimming workspace files (AGENTS.md/SOUL.md/MEMORY.md and daily memory), enabling promp... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 4742 次。
如何安装 OpenClaw Token Optimizer？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install openclaw-token-save」即可一键安装，无需额外配置。
OpenClaw Token Optimizer 是免费的吗？
是的，OpenClaw Token Optimizer 完全免费（开源免费），可自由下载、安装和使用。
OpenClaw Token Optimizer 支持哪些平台？
OpenClaw Token Optimizer 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 OpenClaw Token Optimizer？
由 ccjingeth（@ccjingeth）开发并维护，当前版本 v1.0.0。
推荐 Skills