AI Agent Guardrails

Guardrails are the checks that run on an agent's input and output so it doesn't leak data, get hijacked, or run away with your bill. agentFast runs them as a pipeline in the runtime — before the model ever sees the prompt, and again on everything that gets stored.

PII never reaches the model
AI agent guardrails: PII is redacted before the model ever sees the promptEmails, cards and SSNs are replaced with stable per-run placeholders before the prompt reaches the model and before anything is written to a trace. Hardened against zero-width, homoglyph and spacing obfuscation.user inputrefund tojane@acme.comcard4111 1111 1111 1111guardrail pipelinenormalize obfuscationdetect PIIstable placeholderswhat the model seesrefund to[EMAIL_1]card[CARD_1]same value → same placeholder, per run · map never leaves process memory · traces stored redacted
The full pipeline

Obfuscation-hardened PII

Zero-width characters, homoglyph look-alikes and unusual spacing are normalized first, so obfuscated PII is still caught. Came out of an adversarial pass, with regression tests.

Injection neutralization

Instructions smuggled in tool output or retrieved content are flagged or neutralized rather than obeyed.

Abuse blocking

Unsafe requests are refused at the guardrail layer, not left to prompt-wrangling.

Rate limits

Token-bucket limits, global and per-tool, backed by Redis when you have it — so a runaway loop can't exhaust a downstream API.

Retrieval grounding

A confidence score on how well an answer is supported by retrieved context, so you can catch ungrounded claims.

On by config, every SDK

Declared in agentfast.yaml and enforced by the runtime, so they apply the same on LangGraph, CrewAI, or a vanilla loop.

Guardrails pair with observability (every intervention is a visible step) and human-in-the-loop (the last line before an irreversible action). See them on LangGraph or CrewAI.

Stop rebuilding the plumbing.

agentFast is the production layer — memory, 50 tools, observability, evals, guardrails, human-in-the-loop — for LangGraph, CrewAI, the Claude & OpenAI Agent SDKs & Vanilla. Own it for life.

Get agentFast — from $89