-
Architecture fix for long-running tool execution: 1. ClaudeCliProtocol: Two-tier idle timeout - Chat mode: 45s (detect API hang/silent rate limit) - Tool execution mode: 5min (Bash long tasks like pip install) - Automatically switches on tool_use_start / content_block_stop events 2. Provision API: POST /templates/:templateId/provision - Runs shell commands directly via child_process.exec - Bypasses Claude ACP entirely (no idle timeout issue) - 10min per-command timeout, security blocklist - Frontend button with prompt for custom commands 3. Template UI: Added 🔧 安装依赖 button (teal)
-
…aruncreate skill paths - Upload skill files to .claude/skills/ (Claude Code ACP reads from here) - Add conversation type filter (user/callback/task/train) to AgentChat - Parse messages.log for richer conversation history display - Fix idearuncreate template & company agent skill directory structure - Add adm-zip dependency for ZIP extraction - Add fs/path top-level requires in company.js
-
- Agent as Micro-Container: each agent lives in data/agents/{id}/ with full encapsulation - Persona vs Capability split: agent.md (identity) + .claude/skills/main_skill/SKILL.md (workflow) - Conversations co-located: data/agents/{id}/conversations/ replaces global dir - Meta Skills renamed: meta_hunt, meta_digest, meta_creator, meta_evolving, meta_benchmark, meta_research - Dynamic Hot-Reload: AgentManager.getAgentConfig() falls back to filesystem scan - Parallel Sub-Agent orchestration via Promise.all with multi-terminal UI - Removed SQLite dependency, global skills/ directory, and legacy conversations/ - Full architecture docs: 18-local-filesystem-first-agent-encapsulation.md