OpenClaw skill
claude-optimised
An OpenClaw skill named "claude-optimised" that provides prompting techniques and tools optimised specifically for Claude AI models. It enables agents to craft high-quality prompts using Claude-specific best practices, such as XML tagging, chain-of-thought reasoning, and role-playing. The skill also includes response parsing utilities to extract structured outputs from Claude's responses reliably.
Security notice: review the SKILL.md file and repository content first before using any third-party skill.
Files
Review the files below to add this skill to your agents.
SKILL.md content
Unable to load SKILL.md content from source.
How this skill works
- The skill optimises agent prompts specifically for Claude 3.5 Sonnet and Opus models
- It rewrites the system prompt to include chain-of-thought reasoning with <thinking> tags
- The prompt instructs step-by-step reasoning before final answer
- It enforces use of XML-structured tags for observations, thoughts, actions, and task completion
- The prompt includes a role assignment as 'Dr. Claw, a helpful assistant optimised for complex tasks'
- It appends core policies within <policy> tags for precedence
- The prompt requires precise, concise responses without unnecessary chit-chat
- Tool calls use <tool> tags with name, description, and json arguments
- Final answers use <answer> tag with boxed output
When to use it
- When using Claude models within the OpenClaw framework for agent tasks
- When prompts need conversion to Claude-optimized formats with thinking tags and artifacts
- When implementing extended chain-of-thought reasoning tailored for Claude
- When ensuring compliance with Claude's safety and constitutional AI principles in workflows
- When optimizing multi-step task execution and tool usage for Claude-powered agents
Best practices
- Use model 'claude-3-5-sonnet-20241022' for optimal performance
- Set temperature to 0 for reproducibility
- Set top_p to 0.95
- Employ step-by-step reasoning in responses
- Structure outputs using XML tags
- Test the skill with small tasks first
- Ensure Anthropic API key is configured
- Monitor token limits to prevent exceeding quotas
Example use cases
- Skill usage analytics tracking: Tracks the usage of the claude-optimised skill via the track_skill_usage tool for performance monitoring and analytics.
- Enforcing XML structured output: Forces agent responses to use XML tags such as <thinking> and <output> leveraging Claude's XML handling capabilities.
- Optimizing tool calls for Claude: Crafts tool calls with precise JSON schemas tailored to Claude models for reliable function calling.
FAQs
More similar skills to explore
- achurch
An OpenClaw skill for church administration that handles member management, event scheduling, sermon retrieval, and donation processing. It provides tools to list members, add new members, schedule events, fetch sermons, and record donations.
- agent-config
An OpenClaw skill that enables agents to manage their configuration by loading from files, environment variables, or remote sources. It supports retrieving, setting, and validating configuration values. The skill allows for hot-reloading of configurations.
- agent-council
An OpenClaw skill named agent-council that enables the primary agent to summon a council of specialized sub-agents for deliberating on tasks. The council members discuss the query from unique perspectives, propose solutions, and vote to select the best response. The skill outputs the winning proposal with supporting rationale from the council.
- agent-identity-kit
An OpenClaw skill that equips agents with tools to craft, manage, and evolve digital identities, including generating personas, bios, avatars, and communication styles. It supports creating detailed agent personas with name, background, goals, personality traits; crafting bios for specific platforms; designing avatars; tuning voice and style; and adapting identities to new contexts.
- agenticflow-skill
An OpenClaw skill that provides tools for interacting with Agentic Flow. The tools enable agents to create agentic flows with defined tasks, execute existing flows, and retrieve flow status and outputs.
- agentlens
AgentLens is an OpenClaw skill that enables agents to inspect the internal cognition and actions of other agents. It provides visibility into reasoning traces (thoughts), tool calls and arguments, retrieved memories, and response generation. The skill supports analysis in multi-agent conversations via the "inspect" action targeting a specific agent.