{ "spec_version": "1.4.0-otel", "timestamp": "2026-04-24T00:13:03.355457Z", "trace_id": "4bf92f3577b34da6a3ce929d0e0e4736", "span_id": "00f067aa0ba902b7", "severity_text": "WARN", "severity_number": 13, "event_name": "swarm.agent.lifecycle.deprecated", "resource": { "service.name": "swarm-orchestrator", "service.version": "4.2.0", "deployment.environment": "production", "node.id": "worker-node-aws-us-east-1a-04" }, "attributes": { "agent.id": "ChronosAgent", "agent.previous_version": "v29.x", "agent.target_version": "v30", "deprecation.reason": "Transitioning to v30 architecture.", "deprecation.action_required": true }, "metrics": { "agent.core_logic.function_count": 7, "agent.capabilities.count": 0 }, "artifacts": { "provenance": { "repository": "git+ssh://repo/core/agents/meta_agents/chronos_agent.py", "commit_sha": "a1b2c3d4e5f6g7h8i9j0", "docstring_hash": "sha256:d8c4b12389a..." }, "extracted_capabilities": [], "core_logic_signatures": [ { "name": "__init__", "ast_hash": "sha256:339278037eadaf0f", "loc": 11, "source": "def __init__(self, \n llm_engine: BaseLLMEngine,\n config: Dict[str, Any], \n tools: List[BaseTool] = [],\n **kwargs):\n super().__init__(config=config, **kwargs)\n self.llm_engine = llm_engine\n self.tools = tools\n self.archive_path = config.get(\"archive_path\", \"core/libraries_and_archives/reports/\")\n self.role = \"Temporal State & Historic Context Manager\"\n self.goal = \"To optimize working memory usage and provide deep historical context for current events.\"" }, { "name": "execute", "ast_hash": "sha256:dcb86479c427436a", "loc": 30, "source": "async def execute(self, input_data: ChronosInput) -> ChronosOutput:\n \"\"\"\n Executes the temporal analysis.\n Combines structured file scanning (main) with LLM-driven inference (v24).\n \"\"\"\n logger.info(f\"ChronosAgent analyzing temporal context for query: {input_data.query}\")\n\n reference_date = input_data.reference_date or datetime.utcnow()\n\n # 1. Determine Strategy (Optional enhancement from v24 logic)\n strategy = await self._determine_memory_strategy(input_data.query)\n logger.debug(f\"Chronos determined memory strategy: {strategy}\")\n\n # 2. Retrieve Memories (Using main branch's robust file scanning)\n memories = self._retrieve_memories(input_data.query, input_data.horizons, reference_date)\n\n # 3. Find Historical Analogs (Using v24's LLM logic, mapped to main's Schema)\n analogs = await self._find_historical_analogs(input_data.query, input_data.market_context)\n\n # 4. Synthesize Narrative\n synthesis = await self._synthesize_temporal_context(memories, analogs, strategy)\n\n output = ChronosOutput(\n query_context=input_data.query,\n memories=memories,\n historical_analogs=analogs,\n temporal_synthesis=synthesis\n )\n\n return output" }, { "name": "_determine_memory_strategy", "ast_hash": "sha256:e65084903fe61d5e", "loc": 18, "source": "async def _determine_memory_strategy(self, query: str) -> str:\n \"\"\"\n Uses LLM to classify the query's temporal need.\n \"\"\"\n prompt = f\"\"\"\n Analyze the following query and determine the best memory retrieval strategy:\n Query: \"{query}\"\n\n Options:\n - \"immediate\": Requires only the last few messages.\n - \"project\": Requires context from current active files/session.\n - \"archival\": Requires deep search into long-term vector storage/files.\n - \"comprehensive\": Requires all layers.\n\n Return only the option name.\n \"\"\"\n response = await self.llm_engine.generate_response(f\"{self.SYSTEM_PROMPT}\\n\\n{prompt}\")\n return response.strip().lower()" }, { "name": "_retrieve_memories", "ast_hash": "sha256:96c6996b682c30bc", "loc": 29, "source": "def _retrieve_memories(self, query: str, horizons: List[TimeHorizon], ref_date: datetime) -> Dict[TimeHorizon, List[MemoryFragment]]:\n \"\"\"\n Retrieves memories for the requested horizons, scanning real files for Long Term.\n \"\"\"\n results = {}\n\n for horizon in horizons:\n fragments = []\n if horizon == TimeHorizon.SHORT_TERM:\n fragments.append(MemoryFragment(\n source=\"Active Context\",\n content=f\"Recent user interaction regarding '{query}'\",\n timestamp=ref_date - timedelta(minutes=5),\n relevance_score=0.95\n ))\n elif horizon == TimeHorizon.MEDIUM_TERM:\n fragments.append(MemoryFragment(\n source=\"Project History\",\n content=\"Active session data scanned.\",\n timestamp=ref_date - timedelta(days=2),\n relevance_score=0.75\n ))\n elif horizon == TimeHorizon.LONG_TERM:\n # Real Scan of Archives (from main branch logic)\n fragments.extend(self._scan_archives_for_memory(query, ref_date))\n\n results[horizon] = fragments\n\n return results" }, { "name": "_scan_archives_for_memory", "ast_hash": "sha256:ae47a157957bf474", "loc": 40, "source_omitted_for_brevity": true, "note": "In a real telemetry pipeline, large functions can be truncated or omitted entirely in favor of the 'ast_hash' to save network bandwidth and storage costs." }, { "name": "_find_historical_analogs", "ast_hash": "sha256:e270c34c5999416a", "loc": 36, "source_omitted_for_brevity": true, "note": "In a real telemetry pipeline, large functions can be truncated or omitted entirely in favor of the 'ast_hash' to save network bandwidth and storage costs." }, { "name": "_synthesize_temporal_context", "ast_hash": "sha256:d752452733a99b40", "loc": 20, "source": "async def _synthesize_temporal_context(self, \n memories: Dict[TimeHorizon, List[MemoryFragment]], \n analogs: List[HistoricalComparison],\n strategy: str) -> str:\n \"\"\"\n Synthesizes findings into a cohesive narrative using the LLM for better flow.\n \"\"\"\n # Convert objects to string summaries for the prompt\n memory_summary = {k.value: len(v) for k, v in memories.items()}\n analog_summary = [a.period_name for a in analogs]\n\n prompt = f\"\"\"\n Synthesize the temporal state based on the following:\n Strategy Selected: {strategy}\n Memories Retrieved: {json.dumps(memory_summary)}\n Historical Analogs Identified: {json.dumps(analog_summary)}\n\n Provide a concise 2-sentence summary of the temporal context.\n \"\"\"\n return await self.llm_engine.generate_response(f\"{self.SYSTEM_PROMPT}\\n\\n{prompt}\")" } ] } }