RFC 9421 and Open wbauth Protocol Accelerate AI Agents 5.5x by Replacing Anti-Detect Browsers with Cryptographic Signatures

A new open implementation of the IETF RFC 9421 HTTP Message Signatures standard called wbauth allows AI agents to authenticate using Ed25519 cryptographic signatures instead of resource-heavy anti-detect browser farms. The approach eliminates the need for Camoufox, Playwright, or AdsPower instances that consume 500 MB to 1.5 GB of RAM per process and require residential proxy pools. Benchmarks show cold start time dropping from 1744 ms to 5 ms, request time from 86.8 ms to 15.8 ms, and memory usage from 500 MB to 0.76 MB. The system uses three HTTP headers for signing and verification against a public key directory hosted on Cloudflare Workers and D1. Server-side validation at the WAF or edge involves fetching the public key, canonicalizing the message per RFC 9421, verifying the Ed25519 signature, and checking timestamps to prevent replay attacks. The npm package wb-auth@0.4.0 and PyPI package wbauth-sdk==0.4.0 provide ready-to-use signers for Node, Bun, Deno, Python httpx, and requests. The full source is available on GitHub under the MrKoriy/wbauth repository.

HabrTranslated from Russian

Read at source

The typical 2026 AI agent farm relies on clusters of nodes running Camoufox, Playwright, or AdsPower, each consuming 500 MB to 1.5 GB of memory, plus residential IP pools billed by the gigabyte and separate modules that simulate human-like mouse movements with Bézier curves.

This architecture forces organizations into a zero-sum arms race: defenders build increasingly sophisticated fingerprint detection while attackers invest ever more resources in evasion. The result is wasted hardware, rising costs, and fragile systems that break after every WAF heuristic update.

RFC 9421 (HTTP Message Signatures, published February 2024 by the IETF) replaces mimicry with cryptographic proof. An agent signs the request with an Ed25519 private key; the server verifies the signature against a public key retrieved from a trusted directory indicated by the new Signature-Agent header.

The open-source wbauth project implements this standard with three components: the wb-auth@0.4.0 npm package for Node/Bun/Deno/Workers, the wbauth-sdk@0.4.0 PyPI package with httpx and requests adapters, and a Cloudflare Workers + D1 directory for key registration, rotation, and revocation.

Performance comparisons demonstrate dramatic gains: cold start falls from 1744 ms to 5 ms, per-request latency drops from 86.8 ms to 15.8 ms, memory usage shrinks from 500 MB to 0.76 MB, and runtime disk footprint decreases from 1.3 GB to 14 MB. The solution also removes dependency on proxies and eliminates breakage during WAF updates.

Server-side verification at the edge or WAF follows five steps: retrieve the public key from the directory with KV/Redis caching, build the canonical signature string according to RFC 9421 rules, perform the Ed25519 verification, validate created and expires timestamps to block replay attacks, and forward the request to the origin API without triggering CAPTCHA or blocking mechanisms.

Developers can integrate the protocol immediately using the provided code examples for Python and TypeScript. The project invites community discussion on shifting from fingerprint evasion to standardized cryptographic bot authentication.

Meridian digestThe week's most consequential security stories, distilled into one briefing.See the digest

More in AI Security

Logical Argument Claims Full Alignment of Superintelligent AGI Is Fundamentally Impossible
Claude Account Bans Linked to IP Reputation, Device Environment and Payment Channels, But No Confirmed Formula
Local LLMs in AppSec: What Open Tools and Models Can Handle in Application Security