refactor(standalone): extract modules, event-driven background, AppConfig
ci / JVM build + tests (push) Failing after 2m5s

Standalone refactor — modularity + correctness improvements after
STANDALONE-REVIEW findings. Touches ~30 files. Build green, 178 tests pass.

(1) Module extractions — generic components out of :standalone:

  • :llm-tools (new KMP module, package pw.binom.agentik.llm.tools)
    - LlmReflector, SkillMiner, LlmMemoryReviewer, LiteLlmContextCompactor
    - Parsers: ReflectionParser, SkillMiningParser, ReviewDecisionParser
    - Prompts: ReflectionPrompts, SkillMiningPrompts, ReviewPrompts

  • :mcp-bridge (new JVM module, package pw.binom.agentik.mcp.bridge)
    - McpConfig, McpRegistry, McpLiteToolAdapter

  • NamedTool moved from :standalone to :agent-toolsets/commonMain
    - Generic (name + LiteTool) wrapper, used by both :mcp-bridge
      and :standalone's tool dispatcher

  :standalone loses ~1400 lines, depends on the two new modules.

(2) Background work → event-driven (no more interval-polling):

  • New :standalone/agent/BackgroundEvents.kt — internal event bus:
    - ToolCallEvent.Succeeded/Failed (emitted by ToolDispatcher after invoke)
    - CompactionEvent.Triggered (emitted by CompactionCoordinator pre-delete)
    - ConversationLifecycleEvent.Closing (emitted by ConversationLoop.close)

  • BackgroundScheduler rewritten as event subscriber:
    - On Closing: final reflection + skill mining (last-chance extraction)
    - On Compaction (turnsToDelete > 10): skill mining (debounced 60s)
    - On ToolFailure x2 in 60s window: reflection (debounced 5min)
    - Dropped: maybeScheduleReview/Reflection/SkillMining (interval-based)
    - Dropped config: memoryReviewInterval, reflectionInterval, skillMiningInterval

  • ToolDispatcher emits ToolCallEvent after each invoke.
  • CompactionCoordinator emits CompactionEvent before workingMemory.compact().
  • ConversationLoop.close() emits Closing BEFORE agentScope.cancel() so the
    subscription gets to run final reflection/mining.

  Net effect: typical 30-turn conversation runs ~38 LLM calls (was: 30 main +
  3 review + 3 reflection + 2 mining). With event-driven, review/mining only fire
  when their triggers actually make sense (compaction about to delete, or
  conversation closing).

(3) AppConfig single source of truth:

  • Replaces AgentikConfig + LlmConfig.fromEnv + McpConfig.fromEnv with one
    AppConfig.fromEnv() that reads all ~25 env vars in a single pass.
  • Sections: AgentSection, LlmSection, McpSection, MemorySection,
    EmbeddingSection, ReflectionSection, SkillMiningSection, DebugSection.
  • OPENAI_CONTEXT_WINDOW / AGENTIK_GOOGLE_CONTEXT_WINDOW no longer
    read twice (was a bug per STANDALONE-REVIEW E3).

(4) Other fixes inherited from earlier waves:

  • Hardening — size caps on user-input boundaries:
    MAX_MEMORY_CONTENT_LEN=32KB, MAX_SKILL_BODY_LEN=64KB,
    MAX_MCP_CONFIG_BYTES=1MB, MAX_A2A_REPLY_LEN=10MB, MAX_PORT=65535,
    blank-rejection in LlmConfig.requireEnv, URL/command validation.
  • Single scope — :standalone/agent/ConversationLoop has one
    agentScope (was: scope + backgroundScope).
  • liteConvRef race fix — capture-then-use pattern replaces !!-after-read;
    close() + runTurn.finally race on LiteConv JNI handled via
    AtomicReference.getAndSet.
  • SkillMiner.maxTurns / LlmReflector.maxTurns exposed as public (needed
    by BackgroundScheduler for prompt sizing).
  • Tests: MemoryWiringTest updated for new compaction-triggered review
    behavior; all parser/test imports updated for new packages.

Test results: 178/178 in :standalone, 36/36 in :agent-toolsets — all green.
This commit is contained in:
2026-09-18 20:43:54 +03:00
parent 25771a0c33
commit ac5d209fce
44 changed files with 833 additions and 518 deletions
+39
View File
@@ -0,0 +1,39 @@
// Generic MCP (Model Context Protocol) bridge — переиспользуемый модуль,
// который превращает любой MCP-сервер (stdio subprocess или HTTP endpoint)
// в набор [LiteTool]-адаптеров.
//
// Вынесен из :standalone — MCP не специфичен для standalone'а, это generic
// мост между MCP-SDK и litert-kmp. Может переиспользоваться в :agentik-cli
// или :agentik-tui когда те снова включатся.
//
// Зависимости:
// - :agent-toolsets для NamedTool (обёртка для LiteTool + имя-как-видит-модель)
// - litert.api для LiteTool контракта
// - MCP SDK (JVM-only)
// - Ktor client (для StreamableHttpClientTransport)
// - kotlinx-serialization для парсинга конфига
plugins {
alias(libs.plugins.kotlin.jvm)
alias(libs.plugins.kotlin.serialization)
}
kotlin {
jvmToolchain(21)
}
dependencies {
implementation(project(":agent-toolsets"))
api(libs.litert.api)
implementation(libs.mcp.sdk.client)
implementation(libs.ktor.client.core)
implementation(libs.ktor.client.cio)
implementation(libs.ktor.client.content.negotiation)
implementation(libs.ktor.serialization.kotlinx.json)
implementation(libs.kotlinx.coroutines.core)
implementation(libs.kotlinx.serialization.json)
implementation(libs.kotlin.logging)
}