← 返回 Skills 市场
127
总下载
1
收藏
0
当前安装
5
版本数
在 OpenClaw 中安装
/install optimoclaw-skill
功能描述
Analyze OpenClaw configs and session data to recommend specific, cost-effective token usage and model tuning changes with clear trade-offs and explanations.
安全使用建议
This skill is internally coherent for optimizing OpenClaw token usage, but before installing: 1) Confirm you have the openclaw CLI installed and test what `openclaw config get 'agents' --json` and `openclaw status --usage --json` return in a safe environment — ensure they don't leak API keys or other secrets in your setup. 2) Verify the GitHub source referenced in SKILL.md (Marc4211/optimoclaw) matches the published package and inspect the repo if you need provenance. 3) Run the skill first in a non-production or least-privilege environment (or with a dry-run/open review) so recommendations are not applied automatically; manually review any `openclaw config set` commands before executing. 4) Check the provided rate card and model-cost assumptions against your provider invoices/plans — rates may vary. If any of these checks fail or the openclaw outputs include secrets, do not use the skill until the behavior is clarified or adjusted.
功能分析
Type: OpenClaw Skill
Name: optimoclaw-skill
Version: 1.0.4
The OptimoClaw Token Optimizer is a utility skill designed to help users reduce LLM token costs by analyzing OpenClaw configuration and session data. It uses legitimate CLI commands (`openclaw config get` and `openclaw status`) to gather metrics and provides recommendations for configuration adjustments. There is no evidence of data exfiltration, malicious execution, or harmful prompt injection; the skill operates entirely within the context of the OpenClaw environment to provide cost-benefit analysis.
能力评估
Purpose & Capability
Name/description: token/model optimizer for OpenClaw. SKILL.md: explicitly uses the openclaw CLI to read agent configs and usage and produce configuration recommendations. No unrelated binaries, env vars, or packages are requested. Note: SKILL.md contains a GitHub source URL while registry metadata lists source as unknown — a minor provenance/trust gap to verify.
Instruction Scope
Instructions are specific and scoped: run `openclaw config get 'agents' --json` and `openclaw status --usage --json`, inspect the outputs, and produce recommendations; it explicitly says it will not auto-apply changes. This is coherent with the advertised goal. Caveat: the skill expects to read live configuration/session outputs that may include profile names, model IDs, and gateway connection details — the doc asserts API keys/billing tokens are not returned, but you should validate that for your OpenClaw version/environment before running.
Install Mechanism
Instruction-only skill with no install spec and no code files — lowest installation risk. It only requires the existing openclaw CLI to be present on PATH.
Credentials
No environment variables or credentials are requested by the skill. However, it reads local OpenClaw config and status output which can contain gateway connection details and profile names. That access is proportionate for an optimizer but may expose sensitive metadata; validate what `openclaw config get`/`status --usage` returns in your environment.
Persistence & Privilege
Does not request always:true, does not modify other skills, and does not request permanent presence. Default autonomous invocation is allowed (platform default) but the skill's instructions say it will only present `openclaw config set` commands as recommendations rather than applying them.
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 optimoclaw-skill - 安装完成后，直接呼叫该 Skill 的名称或使用
/optimoclaw-skill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
v1.0.4
No functional or file changes detected in version 1.0.4.
- No code or configuration files were altered for this release.
- SKILL.md content is unchanged except for minor formatting and documentation tweaks.
- No user-facing changes or new features introduced.
v1.0.3
No changes detected in this version.
- Version number updated to 1.0.3, but no file changes were made.
- Behavior, documentation, and features remain unchanged.
v1.0.2
No functional or documentation changes in this version.
- Version bump only; no file or documentation changes detected.
- Behavior and optimization logic remain unchanged from previous version.
v1.0.1
Version 1.0.1
- Added explicit usage instructions: clarifies that OpenClaw CLI must be installed and outlines what information is accessed.
- Stated that no automatic config changes are performed—skill only recommends, user reviews and applies.
- Provided additional detail about what data is (and is not) returned by CLI commands (no keys or secrets).
- No changes to optimization logic, rate cards, or settings; documentation improved for security and operational clarity.
v1.0.0
OptimoClaw Token Optimizer v1.0.0 — initial release
- Introduces token optimization guidance for OpenClaw agent systems.
- Provides a detailed rate card for Anthropic and OpenAI models, including local models.
- Outlines the 10 key optimization levers in configuration, with recommended ranges and trade-offs.
- Supplies ready-made optimization profiles: Lean, Balanced, and Quality.
- Includes instructions on analyzing cache ratios and context utilization in session data.
- Defines a structured approach for giving clear, actionable recommendations based on real configuration and usage data.
元数据
常见问题
OptimoClaw 是什么？
Analyze OpenClaw configs and session data to recommend specific, cost-effective token usage and model tuning changes with clear trade-offs and explanations. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 127 次。
如何安装 OptimoClaw？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install optimoclaw-skill」即可一键安装，无需额外配置。
OptimoClaw 是免费的吗？
是的，OptimoClaw 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
OptimoClaw 支持哪些平台？
OptimoClaw 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 OptimoClaw？
由 Marc4211（@marc4211）开发并维护，当前版本 v1.0.4。
推荐 Skills