2 Commits

Author	SHA1	Message	Date
iven	9060935401	perf(runtime): Hermes Phase 1-3 — prompt caching + parallel tools + smart retry ... Phase 1: Anthropic prompt caching - Add cache_control ephemeral on system prompt blocks - Track cache_creation/cache_read tokens in CompletionResponse + StreamChunk Phase 2A: Parallel tool execution - Add ToolConcurrency enum (ReadOnly/Exclusive/Interactive) - JoinSet + Semaphore(3) for bounded parallel tool calls - 7 tools annotated with correct concurrency level - AtomicU32 for lock-free failure tracking in ToolErrorMiddleware Phase 2B: Tool output pruning - prune_tool_outputs() trims old ToolResult > 2000 chars to 500 chars - Integrated into CompactionMiddleware before token estimation Phase 3: Error classification + smart retry - LlmErrorKind + ClassifiedLlmError for structured error mapping - RetryDriver decorator with jittered exponential backoff - Kernel wraps all LLM calls with RetryDriver - CONTEXT_OVERFLOW recovery triggers emergency compaction in loop_runner	2026-04-24 08:39:56 +08:00
iven	504d5746aa	feat(skill-execution): implement execute_skill tool with full execution chain ... - Add ExecuteSkillTool for LLM to call skills during conversation - Implement SkillExecutor trait in Kernel for skill execution - Update AgentLoop to support tool execution with skill_executor - Add default skills_dir configuration in KernelConfig - Connect frontend skillMarketStore to backend skill_list command - Update technical documentation with Skill system architecture Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-03-24 13:24:23 +08:00