← ClaudeAtlas

everme-memorylisted

Use EverMe cloud memory proactively when the user refers to previous conversations, earlier decisions, "last time", "remember when", existing project conventions, or previously solved errors, and save durable user preferences, habits, and decisions the moment they are stated. Do not repeat a search when the host already injected a non-empty <everme_recall> block this turn.
EverMind-AI/EverMe · ★ 81 · AI & Automation · score 82
Install: claude install-skill EverMind-AI/EverMe
# What it gives you When this plugin is loaded, the host has these MCP tools available: | tool | input | output (markdown / JSON) | |-----------------|----------------------------------------------------------------|-------------------------------------------------------------| | `mem_context` | `{ forceRefresh? }` (`query` deprecated/ignored) | markdown Profile snapshot — Profile ONLY, never a search | | `mem_search` | `{ query, topK? }` | markdown search results across all memory buckets | | `mem_save_fact` | `{ fact \| messages, sessionKey?, flush? }` | `{ saved, accepted, status, extracted, profileUpdated, … }` | | `mem_save_turn` | `{ role, text \| messages, sessionKey?, toolCallId?, flush? }` | `{ saved, accepted, status, messageCount, flushed, profileStatus, profileUpdated }` | # Recommended usage **Call these tools autonomously — the moment a trigger fires, not only when the user explicitly asks you to "remember" or "recall".** Each tool's MCP `description` also carries its trigger, so hosts that don't surface the server `instructions` still get the same guidance. **Dedupe with native injection first.** Some hosts inject `<everme_profile>` at session start and `<everme_recall>` before each prompt via native hooks. When a non-empty, relevant block is already in your cont