agentleFS
TakaGotoTakaGoto/rag-learning-academy/.claude/skills/build/SKILL.md

build

Skill24 starsChanged 19 days agoMITView on GitHub

Build a RAG component hands-on with guided steps

168 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use backprop, finances and 9 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • JuliusBrusseeSkill

    backprop

    Bug → spec protocol. When a bug is found or a test fails, trace the cause, decide whether a new §V invariant would catch recurrence, append to §B. This is the one non-obvious thing SDD does that plan-then-execute doesn't. Triggers on test failure, bug report, post-mortem, or explicit user ask.

    Also in 4 projects that use this

  • GerardoRdz96Skill

    finances

    Use when Gera talks about his money — logging an expense ("I spent X on Y"), checking where he stands this pay-period, closing a period, planning debt payoff, onboarding his financial picture, or asking for budgeting/finance opinions. Servy's financial brain. Triggers — "I spent", "log this expense", "where am I this period", "close the period", "how's my budget", "debt plan", "should I pay off", "finances onboard", "/finances".

    Also in 3 projects that use this

  • cpiprintSkill

    favicon

    Generate favicons from a source image

    Also in 3 projects that use this

  • madeehahassanSkill

    Go-Live

    Pre-launch checklist. Use before making any app public, sharing a link externally, or announcing a product launch. Adapts to Claude Code, claude.ai, and Cowork.

    Also in 3 projects that use this

  • JuliusBrusseeSkill

    deepen

    Optional design-improvement pass for when you have spare usage to drain. Finds the shallowest modules in the code the spec touches, researches a deeper design, and proposes refactors that shrink interfaces and hide decisions — behavior held constant, tests green before and after. Proposes §I/§V/§T edits, never silent rewrites. Triggers when the user says "deepen this", "improve the design", "this module feels shallow", "pull complexity down", "use spare budget on the codebase", or invokes /ck:deepen. Leans on the codebase-design skill's deep-module vocabulary when present.

    Also in 3 projects that use this

  • synaptiaiSkill

    gate

    Create a BDSK review_gate artifact. Defines a checkpoint that controls whether generated output may proceed toward acceptance. Use when defining quality, security, or scope gates for execution plans.

    Also in 5 projects that use this

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 5 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 154 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If the user specifies a component (e.g., /build chunker), use that.

    line 16
  2. Very specific:

    Check if a project workspace exists in projects/.

    line 32
  3. Distinctive:

    Create the file structure for the component.

    line 35
  4. Distinctive:

    Keep each stage small and focused.

    line 47
  5. Distinctive:

    Run the tests and verify correctness

    line 61
  6. Very specific:

    Update progress/module-tracker.md to note the completed build exercise.

    line 76
  7. Distinctive:

    If they get stuck, give increasingly specific hints rather than the solution

    line 87
  8. Distinctive:

    Use real-world data examples when possible, not just "lorem ipsum"

    line 90

“Common” is measured: kept by at least 6 in 10 of the 154 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Build: Hands-On RAG Component Construction

Scope: This skill provides step-by-step guided construction of individual RAG components. For open-ended, multi-component challenges that require design decisions, use /challenge.

Guide the learner through building a real RAG component from scratch. This is the core hands-on skill — learners write actual code with step-by-step guidance.

Language awareness: See .claude/LANGUAGE_AWARENESS.md.

Step 1: Choose What to Build

If the user specifies a component (e.g., /build chunker), use that. Otherwise, present the available components based on their track progress:

ComponentDifficultyPrerequisites
Document loaderBeginnerProgramming basics in your chosen language
Text chunkerBeginnerModule 1
Embedding pipelineIntermediateModules 1-2
Vector store setupIntermediateModules 1-3
RetrieverIntermediateModules 1-4
RAG chainIntermediateModules 1-5
Re-rankerAdvancedModules 1-6
Evaluation harnessAdvancedModules 1-7
Full RAG pipelineAdvancedModules 1-8

Step 2: Set Up the Project

  • Check if a project workspace exists in projects/. If not, help the learner create one.
  • Explain the dependencies needed and help install them.
  • Reference templates in .claude/docs/templates/ if available.
  • Create the file structure for the component.

Step 3: Guided Implementation

Walk through the build in clear stages. For each stage:

  1. Explain the goal of this stage — what are we building and why?
  2. Show the interface — what inputs does this piece take, what does it output?
  3. Let the learner write the code — provide the function signature and docstring, let them fill in the implementation.
  4. Review their code — check for correctness, efficiency, and best practices.
  5. Suggest improvements — offer one or two ways to make it better (error handling, performance, configurability).

Keep each stage small and focused. A chunker build might look like:

  • Stage 1: Read a document into text
  • Stage 2: Implement fixed-size chunking
  • Stage 3: Add overlap between chunks
  • Stage 4: Handle edge cases (empty docs, very short docs)
  • Stage 5: Add metadata to chunks

For example, provide a skeleton in the learner's chosen language with the function signature, docstring/comments, and placeholder logic. Use idiomatic patterns for the language.

Step 4: Test the Component

Help the learner write tests for their component:

  • Provide sample input data
  • Define expected outputs
  • Run the tests and verify correctness
  • Discuss edge cases they should think about

Step 5: Integration Check

Explain how this component fits into the larger RAG pipeline. Show how it connects to the previous and next components in the chain. If the learner has already built other components, help them wire this one in.

Step 6: Review and Reflect

Summarize what was built:

  • What the component does
  • Key design decisions made
  • What they learned
  • Ideas for extending it further

Update progress/module-tracker.md to note the completed build exercise.

Suggest 2-3 relevant next steps using slash commands:

  • /evaluate — measure the quality of what you just built with RAG metrics
  • /code-review — get expert feedback on your implementation
  • /challenge — take on an open-ended challenge that combines multiple components

Guidelines

  • Let the learner do the actual coding — guide, do not just provide the answer
  • If they get stuck, give increasingly specific hints rather than the solution
  • Celebrate working code — even simple implementations are achievements
  • Point out common pitfalls specific to each component (e.g., chunk overlap calculations, embedding dimension mismatches)
  • Use real-world data examples when possible, not just "lorem ipsum"

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.