We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
docs: split Tech Agent Design wiki page into 9 sequential sub-pages and create global _Sidebar navigation menu
docs: document 4th recall pipeline (Context Archive Recall) and recovery flow
docs: add concrete messages structure example to context compaction section
docs: fix nested GFM alert rendering issues on wiki
docs: format active recall and fix GFM syntax errors
docs: 更新Agent上下文压缩设计
docs: 补充记忆和工具定义规划
docs: 调整记忆系统设计说明
Updated 11_Tech_Agent_Design (markdown)
docs: document sub-agent fallback policy
docs: clarify memory summary input window
docs: require memory semantic dedup
docs: document memory summary cost controls
docs: align agent runtime wiki docs
docs: clarify memory layer refresh
docs: simplify sub-agent diagram
docs: add sub-agent architecture diagram
docs: clarify sub-agent context budgets
docs: document sub-agent governance
docs: document compaction fidelity safeguards
docs: clarify agent memory conflict handling
docs: document required tool identification & context logic in §6
refactor: deduplicate §5 memory system — merge redundant tables, reorder flowchart, trim trailing dupe
docs: add §5.1 memory write and layer generation — L1 extraction, dedup, L2/L3 distillation
docs: add §5.1 hybrid recall pipeline — FTS5 / code / keyword merge + decay
docs: correct §5.1 cache miss analysis — O(1) miss per round, not zero-cost
docs: add §5.1 transient memory injection and prompt cache impact
docs: update context compaction safety reserve math and table
docs: fix LaTeX percent sign and clarify context compaction formula