skills.rest
mlamla123
mlamla123COMMUNITY
Explore all skills by this author →
GitHub

media-memory

Search indexed images, audio, and PDFs using natural language descriptions.

Updated Aug 23, 2026
Search Tags:#pdf#image#embeddings#audio#semantic-search#chromadb#media-memory
One-click install
npx skills add https://github.com/mlamla123/openclaw-starter-kit --skill media-memory
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: media-memory
Source: https://github.com/mlamla123/openclaw-starter-kit/tree/main/skills/media-memory
Command: npx skills add https://github.com/mlamla123/openclaw-starter-kit --skill media-memory

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Semantic search over all indexed media files (images, voice notes, PDFs) to locate relevant items from past conversations or inbound media, reducing time spent manual sorting and retrieval.

Core Features & Use Cases

  • Semantic search over images, audio, and PDFs using embeddings.
  • Ingest new media from ~/.openclaw/media/inbound/ and index into the memory database for fast retrieval.
  • Use cases include finding a screenshot, a voice note, or a document described in natural language.

Quick Start

Query the media memory with a natural language description to locate the desired media file.

Frequently Asked Questions about media-memory

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I search for an image or audio file using a natural language description?▼

You can perform semantic search over media files by providing a natural language query. The system uses Gemini embeddings with ChromaDB storage to retrieve matching images, audio, and PDFs from your indexed media memory database.

What is the best way to index new PDFs and voice notes for semantic search?▼

To index new media for semantic search, place your files into the local inbound directory. The system ingests files from this folder and indexes them into the media memory database for fast future retrieval using embeddings.

Do I need an API key to use Gemini embeddings for searching media files?▼

Yes, an API key is required to use Gemini embeddings. The semantic search and media ingestion processes require API keys and local workspace configuration to properly store and query data in ChromaDB.

Can I use natural language search to find a specific screenshot or document from past conversations?▼

Yes, you can locate screenshots, voice notes, and PDFs from past conversations or inbound media. Semantic search reduces manual sorting by finding relevant items based on your natural language description.

What are the limitations of using ChromaDB for local media memory storage?▼

ChromaDB storage is limited to local workspace configuration, meaning your media memory database is constrained to your local environment. All indexed files must be placed in the specified local inbound directory for ingestion.

Related Skills

Explore More Data & Analytics Skills→
akselikorhonen-siiliakselikorhonen-siili

data-indexing

Index web pages and local files into ChromaDB using Gemini models.

Community
Intermediate
zrongzrong
11

video-downloader

Download videos from Douyin, WeChat Channels, and yt-dlp-supported sites with JSON metadata sidecars.

Community
Advanced
FutureAtomsFutureAtoms
1

gemini-embeddings

Generate text embeddings via the Gemini API for semantic search and RAG.

Community
Advanced
KARENLHAOKARENLHAO
1

qmd

Index and search local Markdown collections with BM25 keyword search.

Community
Intermediate