← Back to Skills Marketplace
210
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install rag-pipelines
Description
Deep RAG workflow—document ingestion, chunking, metadata, retrieval and reranking, grounding and citations, evaluation, and failure modes (hallucination, sta...
Usage Guidance
This skill is a benign, instruction-only checklist for building and debugging RAG systems. It does not itself install code or ask for credentials, so the immediate risk is low. Before enabling it widely, consider: (1) other skills it references (vector-databases, llm-evaluation) may require API keys or installs—review those separately; (2) the agent running this guidance may have system/network access—ensure the agent's permissions are appropriate, since following the guidance could lead to actions (ingesting documents, connecting to databases) that do require credentials; (3) because it’s instruction-only, behavior depends entirely on the agent and other connected tools—audit those integrations if you plan to run the workflow in production.
Capability Analysis
Type: OpenClaw Skill
Name: rag-pipelines
Version: 1.0.0
The skill bundle contains high-level architectural guidance and a workflow for designing Retrieval-Augmented Generation (RAG) systems. The content in SKILL.md is purely instructional, focusing on best practices for document ingestion, chunking, and evaluation, with no executable code, network calls, or malicious prompt injection attempts.
Capability Assessment
Purpose & Capability
The name/description (deep RAG pipeline) matches the content of SKILL.md. There are no unrelated environment variables, binaries, or install steps required — everything requested is proportional to documenting a workflow.
Instruction Scope
SKILL.md contains guidance and checklists for ingestion, chunking, retrieval, grounding, and evaluation. It does not instruct the agent to read arbitrary files, exfiltrate data, access unspecified env vars, or contact external endpoints. Instructions are scoped to design and debugging of RAG systems.
Install Mechanism
No install specification or code files are present (instruction-only). This is low-risk: nothing is downloaded or written to disk by the skill itself.
Credentials
The skill declares no required environment variables, credentials, or config paths. It does reference other skills (vector-databases, llm-evaluation) which may have their own requirements, but this skill itself does not request secrets.
Persistence & Privilege
always is false and the skill is user-invocable; it does not request persistent system presence or modification of other skills' configurations.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install rag-pipelines - After installation, invoke the skill by name or use
/rag-pipelines - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of the "rag-pipelines" skill, featuring a comprehensive six-stage workflow for building and debugging retrieval-augmented generation (RAG) systems.
- Covers document ingestion, chunking, metadata, retrieval and reranking, grounding with citations, evaluation, and handling of failure modes like hallucination and staleness.
- Includes practical checkpoints, best practices, and a review checklist to ensure robust pipeline construction.
- Provides targeted guidance for debugging and optimizing RAG pipelines, with special notes for handling code and high-stakes domains.
Metadata
Frequently Asked Questions
What is Rag Pipelines?
Deep RAG workflow—document ingestion, chunking, metadata, retrieval and reranking, grounding and citations, evaluation, and failure modes (hallucination, sta... It is an AI Agent Skill for Claude Code / OpenClaw, with 210 downloads so far.
How do I install Rag Pipelines?
Run "/install rag-pipelines" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Rag Pipelines free?
Yes, Rag Pipelines is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Rag Pipelines support?
Rag Pipelines is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Rag Pipelines?
It is built and maintained by ClawKK (@codekungfu); the current version is v1.0.0.
More Skills