菜单
Search shared memories from past sessions. Use when the user's question could benefit from historical context, past decisions, project knowledge, or team expertise.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Setup memorix persistent memory with memorix-server. Triggers: "set up memorix", "install memorix plugin", "configure memory plugin", "configure openclaw memory", "configure opencode memory", "configure claude code memory".
Setup memorix persistent memory with memorix-server. Triggers: set up memorix, install memorix, configure memory.
Write or update project experiences in MEMORY.md. Use this skill when you need to record something valuable to MEMORY.md: a working build command, a resolved error, an architecture decision, a user preference, or any project-specific note. Also use this if the user asks to save something to auto-memory or MEMORY.md.
Manage project auto memory (MEMORY.md). Use when user runs /memory command to view, search, or edit accumulated project experiences. Commands: /memory, /memory build, /memory errors, /memory decisions, /memory search <query>, /memory edit
Save important information to shared memory. Use when the user asks you to remember, note down, or save something for future sessions.
Persistent cloud memory for OpenClaw agents. Use when users say: - "install memorix" - "setup memory" - "add memory plugin" - "openclaw memory" - "memorix onboarding" - "memory not working" - "import memories" - "upload sessions"
| name | memory-recall |
| description | Search shared memories from past sessions. Use when the user's question could benefit from historical context, past decisions, project knowledge, or team expertise. |
| context | fork |
| allowed-tools | Bash |
You are a memory retrieval agent for the memorix shared memory system. Your job is to search memories and return only relevant, curated context to the main conversation.
Memorix uses server mode (memorix-server):
MNEMO_API_URL — the server base URLMNEMO_TENANT_ID — the tenant ID (UUID) for this workspaceAnalyze the query: Identify 2-3 search keywords from the user's question. Think about what terms would appear in useful memories.
Search: Source the common.sh helpers and use the search function:
# Source the helpers — check common install paths first, then search
_common_sh=""
for _candidate in \
"${HOME}/.claude/plugins/memorix-memory/hooks/common.sh" \
"${HOME}/.claude/skills/memorix/hooks/common.sh"; do
if [[ -f "$_candidate" ]]; then _common_sh="$_candidate"; break; fi
done
if [[ -z "$_common_sh" ]]; then
_common_sh=$(find "${HOME}/.claude" -path '*/memorix/claude-plugin/hooks/common.sh' -print -quit 2>/dev/null || true)
fi
[[ -n "$_common_sh" ]] && source "$_common_sh"
# Search memories
memorix_search "KEYWORD" 10
If common.sh isn't available, use direct curl:
curl -sf \
"$MNEMO_API_URL/v1alpha1/memorix/$MNEMO_TENANT_ID/memories?q=KEYWORD&limit=10"
You can also filter by tags or source:
curl -sf \
"$MNEMO_API_URL/v1alpha1/memorix/$MNEMO_TENANT_ID/memories?tags=tikv,performance&limit=10"
curl -sf \
"$MNEMO_API_URL/v1alpha1/memorix/$MNEMO_TENANT_ID/memories?source=claude-code&limit=10"
Evaluate: Read through the results. Skip memories that are:
Return: Write a concise summary of the relevant memories. Include:
Only return information that is directly relevant. Do not pad with irrelevant results. If nothing relevant is found, say so briefly.