SillyTavern Tavern-Card Handler for Openclaw

A powerful utility for parsing, validating, and converting SillyTavern character cards across V1, V2, and V3 specifications.

fanyin1
v1.0.0
Mar 11, 2026
1
1.1k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install charactercard

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install charactercard using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is SillyTavern Tavern-Card Handler?

The tavern-card skill is a specialized tool designed for AI agents within the Openclaw Skills ecosystem to handle SillyTavern character metadata. It provides a bridge between visual character representations and functional AI data, allowing users to interact with complex character personas stored within PNG steganography or structured JSON files.

This skill ensures perfect compatibility with the latest character card standards, enabling agents to extract deeply embedded character traits, world books, and greeting sets. Whether you are migrating old V1 cards to the modern V3 standard or generating new PNG-based cards from scratch, this skill provides the underlying logic to manage these digital assets with technical precision.

SillyTavern Tavern-Card Handler Use Cases

  • Importing character cards directly from PNG images containing embedded metadata.
  • Extracting and displaying detailed character profiles including personality, scenarios, and example dialogues.
  • Converting legacy V1 character card formats into modern V2 or V3 specifications for better model performance.
  • Exporting new AI-generated character personas into standardized SillyTavern PNG files for use in external roleplay applications.
  • Validating character card integrity to ensure all required fields like name and description are present before deployment.

How SillyTavern Tavern-Card Handler Works

  1. The agent identifies a file path (absolute) and uses the read tool to load the raw buffer.
  2. The skill performs a signature check to determine if the file is a standard PNG or a raw JSON object.
  3. For PNG files, the logic iterates through data chunks (tEXt/iTXt) to locate the "chara" keyword and extract the Base64 encoded payload.
  4. Extracted data is normalized into a standard structure, upgrading V1/V2 fields to V3 compatibility where necessary.
  5. The system runs a validation pass to confirm the schema meets the Chara Card Spec requirements.
  6. Finally, the processed character data is returned to the agent for display, interaction, or file generation.

SillyTavern Tavern-Card Handler Setup

To enable this skill within your environment, ensure your workspace is configured to handle binary file reading. No external API keys are required, as the logic is handled locally via the Openclaw Skills framework.

# Ensure your workspace directory is initialized
mkdir -p ~/.openclaw/workspace

# Place your SillyTavern cards in the workspace
cp my_character.png ~/.openclaw/workspace/

SillyTavern Tavern-Card Handler Data Schema & Taxonomy

The skill manages data across three primary versions, organized as follows:

Version Format Key Features
V1 Flat JSON Legacy format with basic fields (name, description, personality).
V2 Nested JSON Standardized spec with data object, creator notes, and tags.
V3 Extended JSON Adds multi-lingual notes, alternate greetings, and asset support (avatars/backgrounds).

Common Data Fields:

  • name: The character's display name.
  • description: Detailed background and physical appearance.
  • personality: Core behavioral traits and psychological profile.
  • character_book: Embedded world-building entries and keyword triggers.

SillyTavern Tavern-Card Handler Advanced Features

  • Steganographic PNG Embedding: High-level support for injecting JSON metadata into PNG tEXt chunks with CRC calculation.
  • Automatic Version Migration: Seamlessly converts V1/V2 data structures into V3 without data loss.
  • World Book Integration: Preserves complex character_book entries, including keyword arrays and insertion orders.
  • Multi-lingual Support: Handles UTF-8 encoding for international character sets and localized creator notes.
  • Base64/JSON Auto-detection: Intelligently handles both raw string and encoded payloads within image chunks.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Requires
Github Stars: 0
forks: 0

Featured*