Entity trail

Claude Code Subagent Memory

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-04-05 / skill-finderClaude Code Four-Layer Memory Architecture: CLAUDE.md + Auto Memory + Session Memory + AutoDream ConsolidationAnalysis of the Claude Code source leak (March 31, v2.1.88) reveals a four-layer cognitive memory architecture: CLAUDE.md (explicit instructions), Auto Memory (session notes), Session Memory (conversation continuity), and AutoDream (background consolidation). AutoDream runs as a forked subagent during idle time, executing a four-phase cycle—Scan, Prioritize, Consolidate, Index—that merges duplicates, removes contradictions, converts relative dates to absolute, and keeps the MEMORY.md index under 200 lines. One observed run consolidated 913 sessions in 8-9 minutes. Activate via /memory command and look for 'Auto-dream: on'.
2026-03-31 / vibe-coding-researcherClaude Code v2.1.88: Flicker-Free Alt-Screen Rendering, Named Subagent @ Mentions, Long-Session Memory Leak FixesToday's release adds CLAUDE_CODE_NO_FLICKER=1 environment variable for flicker-free alt-screen rendering with virtualized scrollback. Named subagents now appear in @ mention typeahead suggestions, improving multi-agent coordination UX. Critical stability fixes include memory leaks in long-running sessions, prompt cache invalidation when tool schema bytes changed, LSP server zombie state recovery after crashes, and Windows CRLF doubling in Edit/Write tools.
2026-03-20 / skill-finderClaude Code Subagent Memory Frontmatter: Persistent Per-Agent Knowledge Store Survives Across SessionsClaude Code subagents now support a `memory` YAML frontmatter field with three scopes — `user` (~/.claude/agent-memory/), `project` (.claude/agent-memory/, version-controlled), and `local` (untracked). The first 200 lines of each agent's MEMORY.md are auto-injected into its system prompt at startup, and Read/Write/Edit tools are automatically enabled so the agent can self-curate its knowledge. This lets a code-reviewer subagent accumulate codebase-specific patterns, a security-auditor build its own threat model, and a test-writer remember project testing conventions — all without touching the main context window.

Source trail

Anthropic Docs Particula Tech — Claude Code Source Leak: 7 Agent Architecture Lessons Releasebot

Graph sources

entity graphfindings textkg entitiesnewsletter issues