Captain Plugin for OpenClaw

Search text, images, video, and audio with natural language using Captain's multimodal RAG API. Index from any source — S3, GCS, Azure, R2, URLs, YouTube, or raw text.

Install

openclaw plugins install @captain-sdk/openclaw-captain

Setup

Get a Captain API key at runcaptain.com/studio, then add to your OpenClaw config:

{
  plugins: {
    entries: {
      captain: {
        apiKey: "cap_...",
        organizationId: "019a..."
      }
    }
  }
}

Tools

Always Available

| Tool | Description | |------|-------------| | captain_search | Search any collection with natural language. Works across text, images, video, and audio. Returns ranked chunks with source citations. | | captain_list_collections | List all collections in your organization with file counts. |

Optional (add to tools.allow)

Collections & Documents:

| Tool | Description | |------|-------------| | captain_create_collection | Create a new collection | | captain_delete_collection | Delete a collection and all its data | | captain_list_documents | List documents in a collection with file names and chunk counts | | captain_delete_document | Delete a specific document by ID | | captain_wipe_documents | Delete all documents in a collection (keeps the collection) |

Jobs:

| Tool | Description | |------|-------------| | captain_job_status | Check indexing job progress, stage, and file counts | | captain_cancel_job | Cancel a running indexing job |

Indexing Sources:

| Tool | Description | |------|-------------| | captain_index_url | Index public URL(s) — documents, web pages, images, video, audio | | captain_index_youtube | Index YouTube video transcripts (up to 20 videos) | | captain_index_text | Index raw text content directly (notes, transcripts, etc.) | | captain_index_s3 | Index from Amazon S3 — bucket, directory, or single file | | captain_index_gcs | Index from Google Cloud Storage — bucket, directory, or file | | captain_index_azure | Index from Azure Blob Storage — container, directory, or file | | captain_index_r2 | Index from Cloudflare R2 — bucket, directory, or file |

To enable all optional tools:

{
  tools: {
    allow: [
      "captain_create_collection",
      "captain_delete_collection",
      "captain_list_documents",
      "captain_delete_document",
      "captain_wipe_documents",
      "captain_job_status",
      "captain_cancel_job",
      "captain_index_url",
      "captain_index_youtube",
      "captain_index_text",
      "captain_index_s3",
      "captain_index_gcs",
      "captain_index_azure",
      "captain_index_r2"
    ]
  }
}

Examples

> Search my-docs for "quarterly revenue trends"
> Search product-catalog for "red sneakers with white sole"
> Search call-recordings for "customer complaints about billing errors"
> Search training-videos for "how to configure SSO settings"
> Index https://example.com/report.pdf into my-docs
> Index this YouTube video into training-videos: https://youtube.com/watch?v=abc123
> Index my S3 bucket "company-data" directory "reports/2026" into financial-docs
> List my collections

Supported File Types

Documents: PDF, DOCX, DOC, TXT, MD, JSON, YAML, CSV, XLSX Images: PNG, JPEG, GIF, BMP, TIFF, WEBP Video: MP4, MOV, AVI, MKV, WEBM Audio: MP3, WAV, AAC, FLAC, M4A, OGG

Links

• Captain Docs
• API Reference
• Multimodal Search
• GitHub