/install aws-knowledge
AWS Knowledge Skill
Query the AWS Knowledge MCP Server via mcporter for real-time AWS expertise.
The MCP server is aws-knowledge (must be configured in mcporter).
Query Routing
Route each question to the right tool:
| Question Type | Tool | Example |
|---|---|---|
| General "how to" / best practices | search_documentation |
"How to secure an S3 bucket" |
| Read a specific doc page | read_documentation |
"Show me the Lambda pricing page" |
| Explore related topics | recommend |
"What else should I know about VPC peering" |
| Is a service available in a region | get_regional_availability |
"Is Bedrock available in ap-southeast-1" |
| List all AWS regions | list_regions |
"What regions does AWS have" |
| Step-by-step task walkthrough | retrieve_agent_sops |
"Walk me through deploying to ECS" |
Tool Calling
All calls go through mcporter:
# Search documentation (most common)
mcporter call aws-knowledge.search_documentation query="\x3Csearch terms>"
# Optionally filter by topic
mcporter call aws-knowledge.search_documentation query="\x3Csearch terms>" topic="\x3Ctopic>"
# Read a specific documentation page
mcporter call aws-knowledge.read_documentation url="\x3Cfull doc URL>"
# Get recommendations for related content
mcporter call aws-knowledge.recommend url="\x3Cdoc URL>"
# List all AWS regions
mcporter call aws-knowledge.list_regions
# Check regional availability
mcporter call aws-knowledge.get_regional_availability service="\x3Cservice name>" region="\x3Cregion id>"
# Get step-by-step SOPs
mcporter call aws-knowledge.retrieve_agent_sops query="\x3Ctask description>"
Workflow
- Classify the question — determine which tool to call (see routing table above)
- Call the tool — use mcporter with appropriate parameters
- Read deeper if needed — use
read_documentationon URLs from search results - Synthesize the answer — combine findings into a clear response
- Include sources — always cite doc URLs so the user can verify
- Offer follow-ups — suggest related topics or deeper dives
Tips
- Search first, don't guess. Even for common topics, search to get the latest guidance.
- Use SOPs for procedures. If the user wants to DO something (deploy, configure, troubleshoot), check
retrieve_agent_sopsfirst — these are tested step-by-step workflows. - Check regional availability before recommending. Don't assume a service is available everywhere.
- Combine tools. A typical flow:
search_documentation→ pick best result →read_documentationfor full content →recommendfor related reading. - Topic filtering. When searching, use the
topicparameter to narrow results if the query is broad.
Domain Patterns
For common query patterns organized by AWS domain, see references/query-patterns.md.
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install aws-knowledge - After installation, invoke the skill by name or use
/aws-knowledge - Provide required inputs per the skill's parameter spec and get structured output
What is Openclaw?
AWS expert powered by the AWS Knowledge MCP Server (via mcporter). Provides real-time access to AWS documentation, best practices, SOPs, and regional availab... It is an AI Agent Skill for Claude Code / OpenClaw, with 106 downloads so far.
How do I install Openclaw?
Run "/install aws-knowledge" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Openclaw free?
Yes, Openclaw is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Openclaw support?
Openclaw is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Openclaw?
It is built and maintained by w0yne (@w0yne); the current version is v1.0.2.