ZeroPrompt.Core 1.1.0

🎯 ZeroPrompt: Sovereign Pure C# Prompt Templating & Grammar-Constrained Engine

ZeroPrompt is an industrial-grade, zero-allocation prompt templating, Pushdown Automaton (PDA) JSON Grammar state machine, Grammar-Constrained Logit Masker, and dynamic few-shot exemplar selector written in 100% pure C#. It operates at Tier 5 (Presentation & Orchestration) of the ZeroPlatform ecosystem.

⚡ Key Capabilities

• Zero-Allocation Prompt Templating (PromptTemplate):
  • Blazing-fast variable interpolation syntax {{variable_name}} without heavyweight regex or runtime reflection.
  • Span-based rendering directly into StringBuilder buffers.
  • Parameter validation and missing variable tracking.
• Pushdown Automaton (PDA) JSON Grammar Engine (JsonGrammarState):
  • Microsecond deterministic stack-based validation of LLM generation streams.
  • Tracks nested objects ({), arrays ([), string literals, escape sequences, colons, and commas.
  • Evaluates allowable next character classes (e.g. AllowOpenBrace, AllowPropertyName, AllowColon, AllowValue, AllowComma).
• Grammar-Constrained Logit Masking (GrammarLogitMasker):
  • Dynamically masks out non-conforming tokens during autoregressive decoding (-Infinity for invalid tokens).
  • Guarantees 100% syntactically valid JSON outputs from local LLMs (ZeroLlm) or remote endpoints, eliminating JSON parse hallucinations in tool-calling workflows.
• Dynamic Few-Shot Exemplar Selector (FewShotSelector):
  • Backed by ZeroVector.Core hardware-accelerated SIMD vector similarity (VectorMetrics.CosineSimilarity).
  • Dynamically picks top-$K$ most semantically relevant few-shot input/output demonstrations for prompts.
• Transformer KV-Cache Layout Optimizer (PromptLayoutOptimizer):
  • Partitions prompt components strictly into cacheable static prefixes (StaticSystem, ToolDefinitions, FewShot, ContextRAG) and dynamic request-specific payloads (History, DynamicSuffix).
  • Guarantees left-to-right prefix stability for local inference engines (ZeroInference / vLLM / llama.cpp), cutting TTFT by up to 70%.
• Prefix Radix Cache (PrefixRadixCache):
  • Sub-microsecond thread-safe prefix tree for caching tokenized prompt prefixes with 64-bit FNV-1a hashing.
• Zero External Dependencies & Multi-Targeting:
  • Compatible with .NET 8.0+, .NET Framework 4.6.2+, and .NET Standard 2.0.

🚀 Quick Start

1. Template Rendering

using ZeroPrompt.Core.Templates;

var template = new PromptTemplate("You are a SCADA diagnostic engineer for line {{line_id}}. Analyze status: {{status}}.");
var context = new PromptContext()
    .Set("line_id", "LINE-04")
    .Set("status", "OVERHEAT_ALARM");

string prompt = template.Render(context);
Console.WriteLine(prompt);

2. JSON Grammar State Machine

using ZeroPrompt.Core.Grammar;

var grammar = new JsonGrammarState();

// Feed tokens as they are sampled from LLM
grammar.Feed("{\"tool\": \"read_register\", \"args\": {\"address\": 40001}}");

if (grammar.IsCompleted)
{
    Console.WriteLine("Valid JSON payload generated!");
}

3. Grammar-Constrained Logit Masking

using ZeroPrompt.Core.Grammar;

var masker = new GrammarLogitMasker();
var grammar = new JsonGrammarState();
grammar.Feed("{\"status\": ");

float[] logits = new float[vocabSize];
string[] vocab = GetVocabularyTokens();

// Masks out tokens that violate JSON syntax rules
masker.ApplyMask(grammar, logits, vocab);

4. Semantic Few-Shot Exemplar Selection

using ZeroPrompt.Core.FewShot;

var selector = new FewShotSelector();
selector.Add(new Exemplar("Read PLC register 100", "{\"tool\":\"read_plc\",\"reg\":100}", embedding1));
selector.Add(new Exemplar("Query temperature sensor", "{\"tool\":\"get_temp\",\"id\":1}", embedding2));

string demonstrationPrompt = selector.BuildDemonstrationPrompt(queryEmbedding, k: 1);

5. Transformer KV-Cache Prompt Layout Optimization

using ZeroPrompt.Core.Caching;

var optimizer = new PromptLayoutOptimizer()
    .AddSystem("You are an autonomous SCADA diagnostic assistant.")
    .AddTools("Tool: read_sensor(name)\nTool: trip_breaker(breaker_id)")
    .AddHistory("User: Check voltage\nAssistant: Voltage is 220V")
    .AddUserQuery("What is the current on Line 02?");

OptimizedPromptLayout layout = optimizer.Optimize();

// Cacheable immutable prefix (System + Tools)
Console.WriteLine($"Prefix Hash: {layout.PrefixHash:X16}");
Console.WriteLine($"Full Prompt:\n{layout.FullPrompt}");

🛡️ Architecture & Tier Compliance

flowchart TD
    PromptTemplate["ZeroPrompt.Core.Templates\n(PromptTemplate, PromptContext)"]
    JsonGrammar["ZeroPrompt.Core.Grammar\n(JsonGrammarState, GrammarLogitMasker)"]
    FewShot["ZeroPrompt.Core.FewShot\n(FewShotSelector, Exemplar)"]
    
    FewShot --> VectorMetrics["ZeroVector.Core.Metrics\n(SIMD CosineSimilarity)"]
    JsonGrammar --> ZeroLlm["ZeroLlm.Core\n(Constrained Sampler)"]
    PromptTemplate --> ZeroAgent["ZeroAgent.Core\n(ReAct Orchestrator)"]

📄 License

MIT License. Engineered with pride for sovereign autonomous computing.