Browse Packages

Built-in input component that collects named input slots and returns structured JSON

Inline resources: LLM, HTTP, Exec, SQL, and Python running before/after main resource

Hybrid expressions mixing get() API calls with arithmetic operators

Advanced HTTP client: auth, retries, custom headers, streaming, and webhooks

File upload handling with multipart form-data support

Single-shot document summarizer that reads a file from --file, stdin, or KDEPS_FILE_PATH and returns a structured summary.

Simple RAG pipeline - index documents and search them with the native embedding executor

All Docker configuration features in kdeps: volumes, networks, env, resources

Online AI-powered CV/JD matching pipeline using cloud LLM APIs

CV/JD matching pipeline using DeepSeek API for LLM and Ollama for embeddings

Fully offline AI-powered CV / Job Description matching pipeline using Ollama for LLM inference.

Control flow with if-else ternary, logical operators, and conditional skipping

Using unpacked component directories alongside .komponent archives

Advanced components: unpacked directories and packed .komponent archives

Component lifecycle hooks: automatic dependency installation and per-run cleanup

Auto-loading .komponent archives from the components/ directory

Workflow driven exclusively by run.component calls (no HTTP server or file reader)

Multi-resource workflow with parallel execution, dependencies, and data flow

ChatGPT-like web UI powered by local LLMs via Ollama

Simple LLM chatbot