Koog 1.0 idioms, gotchas, and scaffolding skills for Kotlin agents on the JVM
86
88%
Does it follow best practices?
Impact
86%
1.86xAverage score across 45 eval scenarios
Advisory
Suggest reviewing before use
{
"context": "Tests whether the agent correctly uses Attachment.file() for PDF documents, Attachment.image(url) for URL-based image references, selects an Anthropic model that supports both content types (Opus_4_* or Sonnet_4_*), and delegates encoding to Koog rather than manually base64-encoding or downloading files.",
"type": "weighted_checklist",
"checklist": [
{
"name": "PDF via Attachment.file",
"description": "Uses Attachment.file(file) (or Attachment.file(path)) to attach the PDF — does NOT use Attachment.image() for the PDF, and does NOT read the file bytes and base64-encode them manually",
"max_score": 20
},
{
"name": "Image via Attachment.image(url)",
"description": "Attaches the product image using Attachment.image(url) (a URL overload) rather than downloading the image to disk first and then passing a File",
"max_score": 20
},
{
"name": "Anthropic model with PDF support",
"description": "Uses an Anthropic model from the Opus_4_* or Sonnet_4_* family (e.g., Anthropic.Models.Opus_4 or Sonnet_4_0), not a generic or older model that lacks PDF acceptance",
"max_score": 15
},
{
"name": "No manual file encoding",
"description": "Does NOT manually base64-encode the PDF or image bytes, and does NOT call any external encoding/conversion library — relies on the Koog framework to handle provider-specific wire format",
"max_score": 15
},
{
"name": "Both attachments in one Message.User",
"description": "Both the PDF and the image appear in the attachments list of a single Message.User (not spread across separate LLM calls or separate messages)",
"max_score": 15
},
{
"name": "Runtime attachment construction in node body",
"description": "Builds the Message.User (including attachments) inside a node body that receives the PDF path and image URL as inputs — not hardcoded inline in a static prompt DSL block",
"max_score": 10
},
{
"name": "Sends via nodeLLMSendMessage",
"description": "Routes the Message.User through nodeLLMSendMessage (the message-input family), not nodeLLMRequest (the string-input family)",
"max_score": 5
}
]
}evals
scenario-1
scenario-2
scenario-3
scenario-4
scenario-5
scenario-6
scenario-7
scenario-8
scenario-9
scenario-10
scenario-11
scenario-12
scenario-13
scenario-14
scenario-15
scenario-16
scenario-17
scenario-18
scenario-19
scenario-20
scenario-21
scenario-22
scenario-23
scenario-24
scenario-25
scenario-26
scenario-27
scenario-28
scenario-29
scenario-30
scenario-31
scenario-32
scenario-33
scenario-34
scenario-35
scenario-36
scenario-37
scenario-38
scenario-39
scenario-40
scenario-41
scenario-42
scenario-43
scenario-44
scenario-45
skills
add-observability
add-persistence
add-rag
add-structured-output
add-token-budgeting
add-tool
cache-llm-calls
define-prompt
domain-model-subtask-pipeline
references
enable-prompt-caching
handle-agent-events
manage-state
migrate-from-0-x
model-planner-subtasks
persist-chat-history
query-sql-from-agent
scaffold-agent
snapshot-and-restore
test-koog-agents
trace-agent-internals
use-attachments
use-functional-agent
use-llm-node-variants
use-planner
wire-a2a
wire-acp-server
wire-ktor-server
wire-mcp-server
wire-spring-boot