← Back to Skills Marketplace
Openclaw Agent Optimize
by
phenomenoner
· GitHub ↗
· v1.2.1
12665
Downloads
50
Stars
122
Active Installs
13
Versions
Install in OpenClaw
/install openclaw-agent-optimize
Description
Use when: you want a structured audit -> options -> recommended plan to improve an OpenClaw workspace (cost, model routing, context discipline, delegation, r...
Usage Guidance
Before installing, expect advisory optimization help rather than automatic changes. Review any proposed config, heartbeat, cron, sub-agent, or memory-file changes, confirm rollback steps, and separately evaluate optional tools such as openclaw-mem before adding them.
Capability Analysis
Type: OpenClaw Skill
Name: openclaw-agent-optimize
Version: 1.2.1
The 'openclaw-agent-optimize' skill bundle is a collection of advisory Markdown documents designed to guide an AI agent in optimizing an OpenClaw workspace for cost, context management, and reliability. It contains no executable code and explicitly enforces a safety contract requiring human approval before any configuration or cron changes are made (SKILL.md, references/safeguards.md). The instructions focus on best practices like tiered model selection, script-first cron jobs, and reducing context bloat without any signs of malicious intent or data exfiltration.
Capability Assessment
Purpose & Capability
The stated purpose is coherent with the content: model routing, context management, cron, heartbeat, delegation, and memory guidance for OpenClaw workspace optimization. Some recommendations could affect automation behavior, but they are presented as proposals rather than automatic actions.
Instruction Scope
The instructions repeatedly require explicit approval before persistent settings, config, or cron changes. The trigger phrase 'agent best practices' is broad, but the skill also scopes itself to OpenClaw optimization and non-immediate-change workflows.
Install Mechanism
The bundle contains Markdown documentation only, with normal ClawHub/manual install instructions and no executable files, install hooks, package code, or hidden scripts.
Credentials
Guidance about isolated cron jobs, memory files, sub-agents, and optional openclaw-mem integration is proportionate to the optimization purpose, but users should review proposed operational changes before approving them.
Persistence & Privilege
The skill does not itself persist or mutate anything, but it may propose persistent config, cron, heartbeat, or memory-file changes; the artifacts require exact changes, rollback plans, verification, and explicit approval first.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install openclaw-agent-optimize - After installation, invoke the skill by name or use
/openclaw-agent-optimize - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.2.1
Add ambient-skill reduction guidance, authoritative context metrics, and verify/rollback-first optimization workflow.
v1.2.0
Polish GTM + portability: clearer advisory contract, add license, tighten wording, and improve quick-start copy.
v1.1.1
Generalize docs: use npx clawhub install; remove non-English text; clarify cross-platform skill directory defaults and scan delay.
v1.1.0
GTM update: add Quick Start prompts, sample output shape, and ops hygiene checklist (snapshot/heartbeat/ClawHub auth).
v1.0.9
Add OpenClaw 2.9+ skills/context notes (snapshot, gating, sandbox env), and update context-clean-up reference to audit-only.
v1.0.8
Add explicit recommendation to use context-clean-up session-store hygiene (report-first, backup-first) when stale sessions accumulate.
v1.0.7
Safety hardening: user-gated config/cron mutations, advisory-first flow, rollback-plan requirement.
v1.0.6
Docs: explain native heartbeat can be costly and not reliably isolatable in practice; recommend disabling native heartbeat and using isolated cron heartbeat + openclaw-mem pairing.
v1.0.4
Docs: recommend native heartbeat session isolation; promote openclaw-mem pairing; clarify canonical ClawHub page + review (審核中) visibility.
v1.0.3
Fixed metadata; Added credits and heartbeat optimization
v1.0.2
Restoring skill; Added credits and heartbeat optimization
v1.0.1
Expanded optimization references and introduced new best practices.
- Added guidance on heartbeat optimization, including control-plane focus and profiling options.
- Introduced safeguards section covering anti-loop and budget guardrails.
- Expanded workflow to highlight heartbeat and user choice for check removals.
- Added four new references: heartbeat-optimization, optimization-playbook, safeguards, and updated README.
- Updated existing references to align with enhanced optimization and safeguard strategies.
v1.0.0
Initial publish
Metadata
Frequently Asked Questions
What is Openclaw Agent Optimize?
Use when: you want a structured audit -> options -> recommended plan to improve an OpenClaw workspace (cost, model routing, context discipline, delegation, r... It is an AI Agent Skill for Claude Code / OpenClaw, with 12665 downloads so far.
How do I install Openclaw Agent Optimize?
Run "/install openclaw-agent-optimize" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Openclaw Agent Optimize free?
Yes, Openclaw Agent Optimize is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Openclaw Agent Optimize support?
Openclaw Agent Optimize is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Openclaw Agent Optimize?
It is built and maintained by phenomenoner (@phenomenoner); the current version is v1.2.1.
More Skills