clojure
π―SkillInstallation
npx skills add g1joshi/agent-skills --skill clojureRun it in a terminal; it asks which AI coding tools to install it for (Claude Code, Codex, Cursor and more).
Uses skills.sh Β· Node.js 22.20+
After installing
Ask for related work and the agent uses it on its own, or call it yourself.
/clojureInstall by hand
Copy the whole skill folder from GitHub to the place below. Copying only SKILL.md can leave out files the skill uses.
Open the skill folder β- This project
.claude/skills/clojure/ - All projects
~/.claude/skills/clojure/
These paths are for Claude Code. For other tools, the command above puts it in the right place.
SKILL.md
View on GitHub βExpert Clojure programming assistance covering immutable data structures, macros, software transactional memory (STM), and REPL workflows. Use when building concurrent functional systems or data transformation pipelines.
Clojure
A Lisp hosted on the JVM (and JS via ClojureScript) with a focus on immutability.
When to Use
- Functional Programming on the JVM: Leveraging immutable data structures, pure functions, and the vast Java library ecosystem.
- Data-Driven & Event-Sourced Systems: Processing complex financial, logistics, and transactional data structures effortlessly.
- Full-Stack Clojure / ClojureScript: Sharing code, validation schemas (Malli/Spec), and business models between backend and browser.
- Interactive REPL-Driven Development: Iterating and hot-reloading code live in running production-like JVM processes.
Quick Start
(println "Hello, World!")
(defn square [x]
(* x x))
(map square [1 2 3]) ; (1 4 9)
Core Concepts
Persistent Immutable Data Structures
Clojure vectors, maps, and sets use structural sharing to provide immutable snapshots with O(log32 N) performance:
;; Creating and updating immutable maps via structural sharing
(def original-user {:id 415 :name "Jane" :roles #{:member}})
(def updated-user (update original-user :roles conj :admin))
;; original-user remains completely unchanged:
;; {:id 415 :name "Jane" :roles #{:member}}
Concurrency Primitives (Atoms & Software Transactional Memory)
Manages shared state transitions safely without manual lock synchronization:
;; Atoms provide atomic, lock-free state updates via compare-and-swap
(def request-count (atom 0))
;; Thread-safe atomic increment
(swap! request-count inc)
(println "Current requests:" @request-count)
Threading Macros (-> and ->>)
Pipelines data transformations cleanly without deeply nested function calls:
(defn calculate-revenue [orders]
(->> orders
(filter :completed?)
(map :total-cents)
(reduce +)
(* 0.01)))
Common Patterns
Threading Macros for Data Transformation
Problem: Deeply nested function calls are hard to read and mentally trace backwards.
Solution:
Use ->> (thread-last) macro for sequential transformations:
(defn process-orders [orders]
(->> orders
(filter #(= (:status %) "completed"))
(map :amount)
(reduce + 0)))
(println (process-orders [{:status "completed" :amount 50}
{:status "pending" :amount 20}
{:status "completed" :amount 30}]))
;; => 80
Best Practices
Do:
- Embrace REPL-Driven Development: Connect your editor (Calva, CIDER) to a running nREPL and test functions interactively.
- Validate Data with Malli or Clojure Spec: Define explicit schemas for function inputs and API boundaries.
- Use Destructuring Comprehensively: Unpack nested maps and vectors cleanly in function argument vectors.
- Leverage Java Interop Directly: Invoke Java classes (
(java.time.Instant/now)) directly without wrapper overhead.
Don't:
- Fight immutability: Do not introduce mutable Java arrays or global variables where immutable maps suffice.
- Hold references to lazy sequences: Realizing unbounded lazy sequences while retaining their head exhausts JVM heap memory.
- Create deep monolithic namespaces: Break namespaces into cohesive, focused modules.
Troubleshooting
| Error | Cause | Solution |
|---|---|---|
CompilerException java.lang.ClassCastException | Passing wrong data structure (e.g. sequence instead of map). | Verify types using (type x) in the REPL. |
ArityException: Wrong number of args | Function called with more or fewer arguments than defined. | Check function parameter vector and multi-arity definitions. |
OutOfMemoryError: Java heap space with lazy sequences | Retaining the head of an unbounded lazy sequence. | Use dorun or doseq to consume lazy seqs without retaining references. |
References
More from this repository10
A template skill providing a starter framework for creating new agent skills with structured sections for triggers, quick start examples, and capability descriptions.
Skill
Skill
Skill
Skill
Skill
Skill
Skill
Skill
Skill