agentsclimarketplace
Sign in

Build rag search stacks

Skill hiteshbandhu/skills-i-use/skills/ai-engineer-talks/build-rag-search-stacks

Runs checklists and workflows for designing, layering, evaluating, and operating production RAG and AI search — hybrid retrieval, agentic context tools, enterprise eval, web neural search, managed platforms, document agents. Use when the user builds retrieval for agents, picks vector vs lexical vs hybrid, designs eval for augmented AI, scales regulated RAG, compares search APIs, or says "RAG stack", "hybrid search", "agentic RAG", "vector benchmark".From its SKILL.md

Install
npx -y skills add hiteshbandhu/skills-i-use --skill build-rag-search-stacks

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

3.6 KB, 822 tokens by cl100k_base, as published. Nobody here has run it

Build RAG and AI search stacks

Action playbook from thirteen AI Engineer talks. Do not summarize talks — pick a workflow and execute it.

Supporting files (read when needed):

  • workflows.md — workflows A–L (steps, deliverables, stop conditions)
  • source-index.md — src-NNN → talk learnings in ingest-into-skills

Optional deliverables: {SKILL_OUTPUT_DIR}/build-rag-search-stacks/ — see skills-i-use OUTPUT.md.

Step 0 — Pick workflow

Use the decision tree below. Open the matching section in workflows.md.

What is the user trying to do?
├─ Choose RAG vs long-context vs fine-tune           → A
├─ Design hybrid retrieval (lexical + vector + rerank) → B
├─ Vet vector DB / ANN vendor claims                 → C
├─ Ship regulated / legal / enterprise corpus RAG      → D
├─ Layer techniques without over-building              → E
├─ Build agent tools for multi-source context          → F
├─ Persist memory + tool scale for agents              → G
├─ Evaluate web / augmented AI search in production    → H
├─ Open-web research agent (long semantic queries)     → I
├─ Fast enterprise doc Q&A (tables, grounding)         → J
├─ Document-heavy knowledge work (Excel, SharePoint)   → K
└─ Vertical product KB (sales, onboarding assets)      → L

Stop summarizing once a workflow is identified — run its checklist.

Install

Copy this folder into your agent’s skills directory, for example:

cp -r skills/ai-engineer-talks/build-rag-search-stacks ~/.claude/skills/
cp -r skills/ai-engineer-talks/build-rag-search-stacks ~/.cursor/skills/
cp -r skills/ai-engineer-talks/build-rag-search-stacks ~/.codex/skills/

From skills-i-use or ingest-into-skills after sync.

Source corpus: ingest-into-skills playlists/rag-search-2025/.

Cross-cutting rules

RuleSource
Embeddings are the dominant lever in most stacks[src-009 @ 4:11]
AI search = concept retrieval + intent + filters + agent loops[src-009 @ 2:27]
Instruction-tuned / steerable embeddings for query shape[src-009 @ 10:37]
Agentic loop: decompose → search → LLM expand → search again[src-009 @ 7:25]
Contextual retrieval: enrich chunks with titles/global metadata[src-010 @ 8:29]
Multimodal: screenshot → multimodal embedding for slides/tables[src-010 @ 15:07]

Disputed steps: read talk in source-index.md.

Output to user

  1. Name the workflow (A–L) and what you are producing
  2. Save artifacts under ./skill-outputs/build-rag-search-stacks/ when the user wants files (diagrams, rubrics, checklists)
  3. Do not auto-commit

Invocation examples

@build-rag-search-stacks design hybrid retrieval for our legal corpus
help me evaluate web search APIs for agents
we need enterprise RAG — where do we start?

What ships with it: 3 files

14.3 KB alongside SKILL.md

Gives 0 of the 12 instructions most evals benchmarks skills give in 822 tokens

Counted across 499 of the 513 authors here whose files we hold, read 2026-09-06

  • Spawn with-skill and baseline runs in the same turnin 31 of 499, across 24 files
  • Keep SKILL.md under 500 linesin 31 of 499, across 24 files
  • Draft assertions while test runs are in progressin 31 of 499, across 24 files
  • Compare against the baseline after changesin 31 of 499, across 13 files
  • Define evals before codingin 26 of 499, across 17 files
  • Run evals frequently during developmentin 25 of 499, across 16 files
  • Keep evals fastin 24 of 499, across 15 files
  • Version evals with codein 24 of 499, across 15 files
  • Generate the eval viewer before evaluating outputs yourselfin 24 of 499, across 17 files
  • Generate an eval report after runsin 24 of 499, across 15 files
  • Track pass@k metrics over timein 22 of 499, across 14 files
  • Save a baseline before making changesin 21 of 499, across 9 files

Said here and by no other author read

  • Pick a workflow and execute its checklist
  • Open the matching section in workflows.md
  • Read source-index.md for disputed steps
  • Name the chosen workflow and what you are producing
  • Save artifacts when the user wants files
  • Treat embeddings as the dominant lever

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.