rag-architect · borghei — Claude Code skill: what it does, how to install
rag-architect — a skill for Claude Code and Cursor designs RAG pipelines and tunes chunking configurations. It benchmarks retrieval quality using Recall@K and MRR on real queries.
LicenseRef-MIT-Commons-Clause 137 installs ★ 1,183 updated 2026-11-08
npx skills add borghei/Claude-Skills --skill rag-architect -a claude-codeWhat rag-architect does
The skill architectures retrieval-augmented generation setups and inspects why existing systems fail to return relevant context. Three bundled CLI scripts drive the workflow: rag_pipeline_designer.py drafts the component layout with cost estimates, chunking_optimizer.py tests 12 chunking modes on corpus samples, and retrieval_evaluator.py scores query matches against TF-IDF baselines.
Validation halts without a ground truth dataset of 50 or more real query-document pairs. Running the evaluation script also requires local Python 3 execution directly in the skill directory.
What it helps with
- design an end-to-end RAG architecture with cost estimates
- pick optimal chunking parameters for a specific corpus
- evaluate embedding models against a TF-IDF baseline
- diagnose retrieval failures using Recall@K and MRR metrics
Example prompts for the agent
- Analyze the document corpus in assets/corpus and recommend chunking settings
- Draft a RAG architecture from requirements.json and calculate monthly costs
- Run retrieval evaluation on sample_queries.json against ground truth
How to install rag-architect
1Skills CLI (Vercel)
npx skills add borghei/Claude-Skills --skill rag-architect -a claude-code2Manually: put the files in place
~/.claude/skills/rag-architect/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/rag-architect/
1Skills CLI (Vercel)
npx skills add borghei/Claude-Skills --skill rag-architect -a codex2Manually: put the files in place
.agents/skills/rag-architect/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add borghei/Claude-Skills --skill rag-architect -a cursor2Manually: put the files in place
.cursor/skills/rag-architect/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install borghei/Claude-Skills engineering/rag-architect/SKILL.md2Manually: put the files in place
.github/skills/rag-architect/SKILL.md1Skills CLI (Vercel)
npx skills add borghei/Claude-Skills --skill rag-architect -a gemini-cli2Platform CLI
gemini skills install https://github.com/borghei/Claude-Skills --path engineering/rag-architectManually: put the files in place
.devin/skills/rag-architect/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add borghei/Claude-Skills --skill rag-architect -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about rag-architect
How do I install rag-architect in Claude Code?
Run npx skills add borghei/Claude-Skills --skill rag-architect -a claude-code. The skill file is placed at ~/.claude/skills/rag-architect/SKILL.md.
What does rag-architect do?
It guides RAG pipeline design, vector database selection, and chunking tuning based on empirical retrieval benchmarks.
What CLI tools are included in the skill?
It includes rag_pipeline_designer.py for pipeline planning, chunking_optimizer.py for text segmentation tests, and retrieval_evaluator.py for scoring search quality.
Similar skills
All skills for AI and ML →rag-architect by other authors comes first.