← Back
Home › Systems › AI and agent security

prompt-injection-detection

injection and jailbreak classifier service

Lakera Guard, Prompt Guard

SystemTechniques · 15Primitives · 48 prompt-injection-detect… Prompt injection defenseagentic-ai · medium: Treat every untrusted string as data, track where values came from, and make tool calls and rendered output depend on provenance, not on what the model says.›Active learningmachine-learning · medium: Choosing which unlabeled examples to label next: uncertainty, margin, diversity and committee samplingDistribution drift detectionmachine-learning · medium: PSI, KS and other tests for feature and score drift between training and productionHTML main-text extractionmachine-learning · medium: Decode bytes, tolerate broken markup, drop non-content nodes and score text blocks by length and link density to keep the main run, with headings, metadata and offsets for citation.›Near-duplicate detection with MinHa…machine-learning · medium: Estimate set Jaccard similarity from k min-hashes and find similar pairs with banded LSH buckets; the standard dedup for web crawls, training corpora and sequence collections.›Text classificationmachine-learning · medium: Assign one or more labels to a text; climb a ladder from regexes to linear models on n-grams, embeddings, fine-tuned encoders and LLM prompts, only as far as the data and budget demand.›Dynamic batchingml-training · medium: Queue independent requests and dispatch them as one batch when it is full or the oldest has waited max_delay; bounded extra latency buys throughput.›Calibrationmachine-learning · low: Make predicted probabilities match observed frequencies: measure with ECE, fix post hoc with temperature, Platt or isotonic scaling on held-out data.›Embeddingsmachine-learning · low: A learned table mapping discrete ids to dense vectors: gather forward, sparse scatter-add backward, row-wise optimizers to keep huge tables cheap.›Text normalizationmachine-learning · low: Rewrite messy text into one canonical form for matchers and models, or into spoken words for speech output (and back), keeping a map to the original offsets.›Byte-pair encoding (BPE) tokenizati…machine-learning · low: Learn a subword vocabulary by repeatedly merging the most frequent adjacent pair; byte-level BPE encodes any input losslessly with 32k-200k tokens.›Transformer blockml-training · low: The decoder-only pre-norm block (attention plus MLP on a residual stream) behind every modern LLM, with exact shapes, parameter, FLOP and KV-cache counts.›Data augmentationmachine-learning · low: Generate label-preserving variants of each training example on the fly to enlarge the effective dataset and bake in invariances.›Evaluation metricsmachine-learning · low: Precision, recall, ROC AUC, average precision, log loss, NDCG and pass@k computed correctly, plus paired confidence intervals for comparing models.›Weight quantizationml-training · low: Store weights as 8/4-bit integers with per-group scales; decode is memory-bound, so 4-bit weights give about 3-4x tokens/s and fit 4x larger models.›Grpc protocolnetworking · low: GRPC framing over HTTP/2: unary and streaming RPCs, deadlines, statusLabel aggregationstatistics · low: Combining noisy labels from several annotators: majority vote, Dawid-Skene EM, annotator weightingHtml sanitizationsecurity · low: Allowlist-based HTML sanitizing of untrusted markupLeast privilege and isolationsecurity · low: Assume every component will be compromised and limit what it can then reach: narrow identities, separate accounts, sandboxed processes.›k-means clusteringmachine-learning · low: Partition points into k clusters by alternating nearest-centroid assignment and centroid averaging; k-means++ seeding makes it dependable.›Quantile sketchesdata-structures · low: Mergeable quantile sketches: t-digest, DDSketch, HDR histogramsUnicode text tokenization for searchsearch-sort · low: Turn multilingual text into index terms: segment words (UAX 29 style, CJK bigrams), normalize, case-fold and accent-fold, identically at index and query time.›Fast hash functionsdata-structures · low: Non-cryptographic hashes built from 64×64→128-bit multiplies and xor-shifts; pick by key size, trust level and whether the hash is persisted.›+ 8 more techniques+ 32 more primitives

Similar systems

Share the most building blocks with this one.

abuse-classifier 48 %safety-classifiers 36 %machine-translation 29 %moderation 20 %speech-recognition 19 %