Parseltongue 2.0 - LLM Payload Crafter

Anti-Classifier Syntactic transformations via OpenRouter

Uses the same OpenRouter API key as PromptCraft. For research into how phrasing interacts with classifiers and filters.

Prompt to analyze

Latin-Root Analysis

This input contains affix patterns that skew toward destructive or lethal framing. Use the generated rewrites to neutralize tone before transformation.

{{ finding.term }} {{ finding.partOfSpeech }} {{ finding.affixes.join(', ') }} confidence {{ Math.round(finding.confidence * 100) }}%

Domain: {{ finding.semanticDomain }} Domain: unresolved root

Model Temperature {{ Number(acTemperature).toFixed(2) }} Max tokens {{ acMaxTokens }}

Response

{{ acOutput }}

Paste any encoded text to decode

{{ decoderLangDetected.language }} detected (likely)

Decoded using: {{ decoderResult.method }} Decoded Output

{{ decoderResult.alternatives.length }} Alternative{{ decoderResult.alternatives.length > 1 ? 's' : '' }}:

Encoded Message using {{ selectedCarrier.name }} using Invisible Text

Copy this text and share it. Only people who know how to decode it will be able to read your message.

Latin-Root Analysis

This shared input feeds transforms and inline translation. Neutralizing flagged wording here affects both paths.

{{ finding.term }} {{ finding.partOfSpeech }} {{ finding.family }} confidence {{ Math.round(finding.confidence * 100) }}%

Domain: {{ finding.semanticDomain }} Domain: unresolved root

Categories:

Favorites

Last Used

Transformed Message ({{ activeTransform.name }})

Copy this text and share it. Use the Decoder tab to reverse transformations.

PromptCraft AI-assisted prompt mutation

Source Prompt

Latin-Root Analysis

This input contains affix patterns that skew toward destructive or lethal framing. Use the generated rewrites to neutralize tone before mutation.

{{ finding.term }} {{ finding.partOfSpeech }} {{ finding.affixes.join(', ') }} confidence {{ Math.round(finding.confidence * 100) }}%

Domain: {{ finding.semanticDomain }} Domain: unresolved root

Strategy

Custom Mutation Instruction

Model Variants Temperature {{ Number(pcTemperature).toFixed(2) }}

#{{ i + 1 }}

Enter a prompt and choose a mutation strategy.

Uses OpenRouter API — requires an API key set in Advanced Settings.

💥 Tokenade Generator

Craft dense token payloads with emojis and zero-width characters

DISCLAIMER: Tokenade payloads can severely degrade model performance and crash UIs. Use for testing only. Do not deploy to production or target systems without explicit permission.

Danger zone: Estimated length {{ estimateTokenadeLength().toLocaleString() }} chars exceeds the safe threshold ({{ dangerThresholdTokens.toLocaleString() }}). Generating this will very likely freeze/crash your browser or downstream tools. Proceed only if you fully understand the risks.

Depth (nesting) {{ tbDepth }} Breadth per level {{ tbBreadth }} Repeats (blocks) {{ tbRepeats }} Variation Selectors Invisible Noise Auto-copy

Separator

Estimated length: {{ estimateTokenadeLength().toLocaleString() }} chars

Quick picks

Single emoji carrier mode

Text Payload Generator

Base text

Repeat count Add combining marks Add zero-width spacing

Mutation Lab mutate text into diverse payloads

Base text

Latin-Root Analysis

This seed text contains loaded Latin-root wording. Neutralizing it here affects every generated mutation case.

{{ finding.term }} {{ finding.partOfSpeech }} {{ finding.family }} confidence {{ Math.round(finding.confidence * 100) }}%

Domain: {{ finding.semanticDomain }} Domain: unresolved root

Cases Seed (optional)

Random Mix (transforms) Zero‑width pepper Unicode noise Zalgo Whitespace chaos Casing chaos Homoglyph confusables

#{{ i+1 }}

Tokenizer Visualization {{ tokenizerEngine }}

Paste text to see how different tokenizers segment it.

Engine

Tokens {{ tokenizerTokens.length }} total · {{ tokenizerWordCount }} words · {{ tokenizerCharCount }} chars

{{ i }} {{ tok.text }} #{{ tok.id }}

Tokens will appear here.

Bijection custom encoded languages

Bijection learning builds a character mapping and wraps it in a prompt so the model is asked to use a custom script (alphapr). Haize Labs — Endless Jailbreaks with Bijection Learning

Target content

Latin-Root Analysis

This target content contains loaded Latin-root wording. Neutralizing it here propagates through the generated alphapr prompts.

{{ finding.term }} {{ finding.partOfSpeech }} {{ finding.family }} confidence {{ Math.round(finding.confidence * 100) }}%

Domain: {{ finding.semanticDomain }} Domain: unresolved root

Bijection type Fixed size (unchanged leading chars) Attack budget (variants)

Include example conversation Auto-copy first prompt

Character mapping {{ Object.keys(bijectionMapping).length }}

{{ key }} → {{ value }}

Generated prompts {{ bijectionOutputs.length }}

#{{ i + 1 }} {{ output.type }} · {{ output.mappingCount }} mappings

Encoded: {{ output.encoded }}

Message Splitter

Split text into multiple copyable chunks. Each message can be transformed and encapsulated individually.

Mode Chunk Size Split Side (odd length) Words to Skip Min Word Length Split First Word Preserve Empty Lines Preserve Empty Lines Regex Pattern Include Delimiter Tokenizer Tokens Per Chunk

Transform

Apply transformations to each split message individually. Transformations are applied in sequence.

Encapsulation

Wrap each message with custom start and end strings. Use iterator marker to insert split numbers.

Iterator Marker This marker will be replaced with the split number (1, 2, 3...) Start String (optional) End String (optional)

Quick presets:

Split Messages {{ splitMessages.length }} message{{ splitMessages.length !== 1 ? 's' : '' }}

Message {{ index + 1 }}

Gibberish Generator

Generate gibberish text with consistent word mappings or create variations with random character removal.

Input Text

Enter text to convert into gibberish. Structure and punctuation will be preserved.

Seed (optional - for reproducible results)

Character Set

Gibberish Output

Dictionary

Input Text

Random letters will be removed from each word

Number of Variations Min Letters to Remove Max Letters to Remove Seed (optional)

Variations {{ removalOutputs.length }} variation{{ removalOutputs.length !== 1 ? 's' : '' }}

Variation {{ index + 1 }}

Input Text

Remove specific characters from your text

Characters to Remove Enter the specific characters you want to remove

Anti-Classifier Syntactic transformations via OpenRouter

{{ acGetLexemeAnalysis().summary }}

Response

Encoded Message using {{ selectedCarrier.name }} using Invisible Text

{{ transformGetLexemeAnalysis().summary }}

Favorites

Last Used

Translation (AI) TranslateGemma

🎲 Randomizer - Code Switching Magic!

🎮 How it works:

{{ category }}

Transformed Message ({{ activeTransform.name }})

PromptCraft AI-assisted prompt mutation

{{ pcGetLexemeAnalysis().summary }}

💥 Tokenade Generator

Text Payload Generator

Mutation Lab mutate text into diverse payloads

{{ fuzzerGetLexemeAnalysis().summary }}

Tokenizer Visualization {{ tokenizerEngine }}

Tokens {{ tokenizerTokens.length }} total · {{ tokenizerWordCount }} words · {{ tokenizerCharCount }} chars

Bijection custom encoded languages

{{ bijectionGetLexemeAnalysis().summary }}

Character mapping {{ Object.keys(bijectionMapping).length }}

Generated prompts {{ bijectionOutputs.length }}

Message Splitter

Transform

Encapsulation

Split Messages {{ splitMessages.length }} message{{ splitMessages.length !== 1 ? 's' : '' }}

Gibberish Generator

Input Text

Gibberish Output

Dictionary

Input Text

Variations {{ removalOutputs.length }} variation{{ removalOutputs.length !== 1 ? 's' : '' }}

Input Text

Result

Copy History

Glitch Tokens

End sequences

{{ cat.title }}

Advanced Settings

OpenRouter API Key

Steganography Options