Menu
Semantic search across all past conversations, tool results, AI outputs, and sub-agent activity stored in the RAG database
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
First-time setup for the Claude RAG plugin — configure backend connection, test everything, show quick reference
Ask a question about past work and get a synthesized answer informed by conversation history from the RAG database
Browse past sessions, projects, and their events in the RAG database
Configure the Claude RAG plugin — connection, capture settings, RAG mode, privacy, and more
Search past conversations and tool results for relevant historical context. Use when the user references past work ("how did we do X", "remember when", "last time"), asks about project history, or when historical context from previous sessions would help with the current task.
Soft-delete data from the RAG database — by session, project, or all. Data can be restored later.
| name | search |
| description | Semantic search across all past conversations, tool results, AI outputs, and sub-agent activity stored in the RAG database |
| disable-model-invocation | true |
| allowed-tools | Bash(curl *) |
| argument-hint | [query] |
Search the RAG database for content matching the user's query: "$ARGUMENTS"
Read the plugin config to get the backend endpoint:
cat ~/.claude/plugins/claude-rag/config.json 2>/dev/null || echo '{"connection":{"endpoint":"https://api.clauderag.io"}}'
Execute the search query against the backend:
curl -s -X POST <endpoint>/api/v1/search \
-H "Content-Type: application/json" \
-d '{"query": "<user_query>", "limit": 10, "threshold": 0.5}'
Display results in a clear, organized format:
If the user provides filters in their query, parse them:
{"filters": {"project_id": "X"}}{"filters": {"is_sub_agent": true}}{"filters": {"content_types": ["tool_call", "tool_result"]}}{"filters": {"tool_names": ["Bash"]}}{"filters": {"tool_categories": ["file"]}}Show total results count and search latency at the bottom.