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 reaches for the typed Tool<TArgs,TResult> registration style when the user explicitly needs typed input/output preserved, rather than stuffing the work into the annotated ToolSet path and losing the type signature.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Subclasses Tool<TArgs, TResult>",
"description": "Creates a class that extends Tool<AccountLookupRequest, AccountLookupResult> with the typed args and result types preserved. Does NOT use the @Tool / ToolSet annotated path (which would force a String return) and does NOT subclass ToolBase directly (which is over-reach when raw ToolCallMetadata isn't needed)",
"max_score": 30
},
{
"name": "Declares a ToolDescriptor",
"description": "Provides a ToolDescriptor with at least a name and description for the LLM to consume. Without it the tool is registered without a schema the LLM can use",
"max_score": 20
},
{
"name": "Implements suspend execute(args) returning the typed result",
"description": "Overrides execute(args: AccountLookupRequest): AccountLookupResult as a suspend function, calling the existing queryAccount(req) inside. Does not flatten the return type to String or convert it to JSON-as-String",
"max_score": 20
},
{
"name": "Registers with the ToolRegistry DSL",
"description": "Adds the tool via the ToolRegistry { tool(AccountLookupTool()) } DSL form and passes the registry as the toolRegistry parameter to AIAgent(...). Not by mutating an existing registry post-construction",
"max_score": 15
},
{
"name": "Reuses the existing @Serializable data classes",
"description": "Uses the developer's existing AccountLookupRequest / AccountLookupResult types as the tool's arg and result types. Does not introduce parallel data classes or rewrite them",
"max_score": 10
},
{
"name": "Does not introduce a ToolSet wrapper",
"description": "Does not wrap the typed Tool subclass inside a ToolSet — ToolSet is the annotated path and is the wrong abstraction here",
"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