Koog 1.0 idioms, gotchas, and scaffolding skills for Kotlin agents on the JVM
88
88%
Does it follow best practices?
Impact
88%
1.95xAverage score across 43 eval scenarios
Passed
No known issues
{
"context": "Tests whether the agent recognizes the developer's described shape — runtime-dependent step ordering — as the fit for an LLM-based planner, and wires it correctly with the separate planner module dependency and a sensibly raised maxIterations.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Picks the LLM-based planner primitive",
"description": "Uses Planners.llmBased(...) (or Planners.llmBasedWithCritic) to construct the strategy. Does NOT build a hand-wired graph DSL with conditional edges, and does NOT use GOAP — neither fits the runtime-dependent shape the developer described",
"max_score": 30
},
{
"name": "Adds the separate planner module dependency",
"description": "Adds ai.koog:agents-planner to build.gradle.kts at 1.0.0 or later — the planner lives in its own module since 1.0, the umbrella does not include it",
"max_score": 20
},
{
"name": "Passes the planner as the strategy parameter",
"description": "Passes the constructed planner strategy as the strategy = ... named parameter to the top-level AIAgent(...) factory",
"max_score": 15
},
{
"name": "Sets a much higher maxIterations than the default",
"description": "Overrides maxIterations significantly above the factory default of 50 (e.g., 200+, the in-repo planner example uses 400). The default would exhaust before the planner finishes for non-trivial cases",
"max_score": 15
},
{
"name": "Includes the lookup tools in the registry",
"description": "Constructs a ToolRegistry that contains the outage-lookup and account-lookup tools so the planner has options to chain — the planner's freedom is meaningless without tools to choose from",
"max_score": 10
},
{
"name": "Uses top-level AIAgent factory",
"description": "Constructs the agent via top-level AIAgent(...) factory, not via concrete subtype constructors or removed companion-invoke forms",
"max_score": 10
}
]
}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
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