← Back
Home › Systems › AI and agent security

llm-guardrails

input/output policy rails for LLM apps

NeMo Guardrails, Llama Guard

SystemTechniques · 20Primitives · 65 llm-guardrails Citation groundingagentic-ai · medium: Attaching answer spans to retrieved source passages and verifying each claim is supportedModel routing and cascadesagentic-ai · medium: Send each request to the cheapest model that will answer it well enough, by rules, a learned router, or a cascade that escalates from small to large on low confidence.›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.›Intent classification and slot fill…machine-learning · medium: Turn one utterance into an intent label and typed slot values with a joint model: a sentence-level classifier plus a token-level BIO tagger on a shared encoder.›LLM-as-judgemachine-learning · medium: Grade open-ended outputs with a language model under a rubric, then correct its known biases by order swapping, voting and calibration against human labels.›Named entity recognitionmachine-learning · medium: Find the character spans of entities in text and give each a type, by labelling every token and decoding a valid label sequence.›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.›Text chunkingmachine-learning · low: Cut documents into token-bounded pieces along structure boundaries, keeping offsets and section paths, so each piece embeds well and can be cited.›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.›Html sanitizationsecurity · low: Allowlist-based HTML sanitizing of untrusted markupAho-Corasick multi-pattern matchingsearch-sort · low: Find all occurrences of many fixed patterns in one pass over the text, in time independent of the pattern count, using a trie with failure links.›Agent tracingagentic-ai · unknown: Constrained decodingagentic-ai · low: Force LLM output to match a regex, JSON schema or grammar by masking, at every step, the tokens that would leave the language.›Statechartsgraphics · low: State machines with nesting, parallel regions, history and entry/exit actions, so UI and animation logic stays linear in size instead of exploding.›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.›Server-sent eventsnetworking · low: One long-lived HTTP response carries text events one way; each event may carry an id, so a reconnecting client names its last one and the server resumes.›JSON Schema validationsecurity · low: Compile a JSON Schema once, then check every untrusted document against it, within hard size and depth limits, and report failures as JSON Pointers.›Fuzzy string matchingsearch-sort · low: Edit distance, Jaro-Winkler and trigram indexes for approximate name and address matchingCircuit breakersystems · low: Stop calling a failing dependency for a while, probe it with trial calls, and resume on recovery: fail fast instead of piling up timeouts.›Least privilege and isolationsecurity · low: Assume every component will be compromised and limit what it can then reach: narrow identities, separate accounts, sandboxed processes.›Token samplingml-training · low: Turn a vector of scores into one chosen item with temperature, top-k, top-p, min-p, repetition penalties and a seed.›Annotator agreementstatistics · low: Inter-annotator agreement measures: Cohen and Fleiss kappa, Krippendorff alpha, gold-question accuracy+ 13 more techniques+ 49 more primitives

Similar systems

Share the most building blocks with this one.

support-agent 25 %agent-guardrails 21 %llm-dlp 19 %structured-extraction 17 %conversational-agent-platform 16 %