Skip to content
Agent Memory
Command

/agent-brain-reset

Output as JSON

BOOST
From plugin
agent-brain
11944 skills9 agents44 commands
Install
> /plugin marketplace add SpillwaveSolutions/agent-brain
> /plugin install agent-brain@agent-brain-marketplace

How it fires

How this command gets triggered: by you, by Claude, or both.

  • Fires itselfClaude auto-loads it when your prompt matches the work.
  • You can call itInvoke it directly when you want it.
  • Slash command/agent-brain-reset

Context preview

What this command does when you run it.

Output as JSON

Command definition

agent-brain-reset.md
name: agent-brain-reset
description: Clear the document index (requires confirmation)
parameters:
  - name: yes
    description: Skip confirmation prompt
    required: false
    default: false
  - name: url
    description: Server URL (default from config or http://127.0.0.1:8000)
    required: false
  - name: json
    description: Output as JSON
    required: false
    default: false
skills:
  - using-agent-brain
last_validated: 2026-03-16

Reset Document Index

Purpose

Clears all indexed documents from the Agent Brain server. This removes all vector embeddings and BM25 index data. Use this when you need to completely rebuild the index or remove outdated content.

Usage

/agent-brain:agent-brain-reset [--yes] [--url <url>] [--json]

Parameters

| Parameter | Required | Default | Description | |-----------|----------|---------|-------------| | --yes, -y | No | false | Skip confirmation prompt and proceed immediately | | --url | No | from config or http://127.0.0.1:8000 | Server URL (env: AGENT_BRAIN_URL) | | --json | No | false | Output as JSON |

Examples

/agent-brain:agent-brain-reset           # Prompts for confirmation
/agent-brain:agent-brain-reset --yes     # Skips confirmation
/agent-brain:agent-brain-reset --json    # JSON output (skips confirmation)

Execution

**IMPORTANT**: This command requires explicit user confirmation before executing.

Step 1: Check Current Index Status

agent-brain status

Report to the user what will be deleted:

  • Number of documents currently indexed
  • Approximate index size
  • Collections that will be cleared

Step 2: Request Confirmation

Before running the reset, you MUST:

1. Show the user what will be deleted 2. Ask for explicit confirmation 3. Only proceed if the user confirms with "yes" or similar affirmative

**Example interaction:**

The following will be permanently deleted:
  - 156 indexed documents
  - 1,247 text chunks
  - Vector embeddings (~45 MB)
  - BM25 index data

Are you sure you want to reset the index? This cannot be undone.

Step 3: Execute Reset

Only after confirmation:

agent-brain reset --yes

Expected Output

Resetting Agent Brain index...

Clearing vector store... done
Clearing BM25 index... done
Clearing metadata... done

Index reset complete.
  Documents removed: 156
  Storage freed: 45.2 MB

The index is now empty. Run '/agent-brain:agent-brain-index <path>' to add documents.

Output

After reset, report:

  • Confirmation that reset completed
  • Number of documents removed
  • Storage space freed
  • Next steps for re-indexing

Error Handling

| Error | Cause | Resolution | |-------|-------|------------| | Server not running | Agent Brain server is stopped | Start with `/agent-brain:agent-brain-start` | | Server Error (409) | Indexing is in progress | Wait for indexing to complete, then retry | | Index already empty | No documents to clear | No action needed | | Permission denied | Cannot write to storage directory | Check directory permissions |

Recovery Commands

# Check server status
agent-brain status

# Start server if needed
agent-brain start

# Verify index is clear
agent-brain status

Safety Notes

  • **This operation is irreversible** - all indexed data is permanently deleted
  • The original source documents are NOT affected
  • Only the search index is cleared
  • Re-indexing will require re-processing all documents
  • Consider backing up the `.agent-brain/` directory before resetting

Notes

  • The reset only affects the current project's index
  • Other project instances are not affected
  • Server remains running after reset
  • Re-index with `/agent-brain:agent-brain-index <path>` after reset
Read more
Ships withagent-brain

A RAG-based (Retrieval-Augmented Generation) document indexing and semantic search system for AI agents and applications. Agent Brain enables intelligent querying of documentation and source code using natural language.

Get the whole plugin

Other commands on agent-brain.