RepoRadar
Leave a Tip
Item detail
github.com

m4xx101/cryptex-oss

RepoRadar surfaced m4xx101/cryptex-oss — a open-source llm red-team lab wit — into the MIT open-source LLM red-team lab with 162 text t section, where it sits at Gold tier with a 'try now' verdict. Its strongest signal is workflow potential, scored 9.3 out of 10.

Open source →Back to radar
Score8.2
Popularity330.0
Risknone
TierGold
Score breakdown
Usefulness8.0
Novelty8.0
Momentum8.0
Maturity8.9
Open-source/build8.4
Evidence8.0
Workflow potential9.3
Setup ease8.8

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for security researchers and AI red-teamers who need a self-hostable adversarial-prompt + encoding lab — Cryptex is the open-source LLM red-team lab with 162 text transformers, 36 mutators, 25 tool surfaces, and 15+ red-team labs covering HarmBench, StrongREJECT, JailbreakBench, indirect injection, glitch tokens, adversarial suffixes, defense fingerprinting, and watermark forensics; for AI

Who should use it

Security researchers and AI red-teamers who need a self-hostable adversarial-prompt + encoding lab — Cryptex is the open-source LLM red-team lab with 162 text transformers, 36 mutators, 25 tool surfaces, and 15+ red-team labs covering HarmBench, StrongREJECT, JailbreakBench, indirect injection, glitch tokens, adversarial suffixes, defense fingerprinting, and watermark forensicsAI safety teams that want a Campaign front door that fans a goal across many attack strategies and judges each with an LLM judge — the Campaign reports a graded ASR you can exportSecurity teams that want a self-hostable browser-based lab with BYOK keys (no backend, no telemetry) — AI calls go from the browser direct to whichever provider you choose, BYOK keys stay in `localStorage`Engineering teams that want to run the same transforms in both SvelteKit (browser) and Python CLI contexts — the canonical transformer list is shared between both surfacesOrganizations that need a private / fully-offline deployment — the multi-arch Docker image (`amd64` + `arm64`) is one `docker run` command, the Dokploy + Traefik + Let's Encrypt path is documented in `DEPLOY.md` for HTTPS VPS deploymentUsers who want a no-install preview — the online demo at `m4xx101.github.io/cryptex-oss/transforms/` shows the full transform surfaceSecurity teams that want a mobile-accessible lab — the responsive web app works in iOS Safari and Chrome Android with all 25 tool surfacesUsers that want a private personal instance reachable from a phone anywhere — the README documents running the Docker image on a Raspberry Pi exposed via Tailscale Funnel or Cloudflare TunnelEvaluation: `docker run -d --name cryptex -p 8080:80 ghcr.io/m4xx101/cryptex-oss:latest`, open `http://localhost:8080`, browse the Transform workbench, run a Campaign, and inspect the graded ASR report

Who should skip it

Skip m4xx101/cryptex-oss if the source repository or demo is inactive, unmaintained, or no longer matches the description shown here.

About this signal

m4xx101/cryptex-oss is tracked by RepoRadar as a open-source llm red-team lab wit in the MIT open-source LLM red-team lab with 162 text t section. It was first seen on 2026-06-25 and last updated on 2026-06-25. The current verdict is 'try now' with a Gold tier and easy setup difficulty. m4xx101/cryptex-oss leads on workflow potential (9.3) and maturity (8.9); its lowest signal is evidence quality (8.0), so factor that in before investing setup time. This page summarizes the evidence RepoRadar has captured from captured source metadata. The score, tier, risk label, and verdict on this page are never influenced by sponsorship, ads, or tips — they reflect only the usefulness, popularity, novelty, momentum, maturity, and evidence signals described in the RepoRadar methodology.

How this item is evaluated

RepoRadar assigned m4xx101/cryptex-oss a composite score of 8.2 out of 10, placing it in the Gold tier. This score combines weighted sub-signals: usefulness (35%), novelty (18%), momentum (14%), maturity (10%), open-source/build quality (7%), evidence quality (6%), workflow potential (6%), and setup ease (4%). Popularity is tracked separately at 330.0 and never affects the composite score or tier. The risk label of 'none' reflects inherent user-impacting hazards, not generic novelty. Items with no risk flag may still require normal code review before production use.

Putting this into practice? Read How to evaluate an AI tool before you adopt it for the checklist behind this score.

Risk explanation

No inherent user-impacting risk is flagged from the captured evidence.

Evidence links

Closest alternatives / related signals

  • No close related item is available in the current public catalog.
cryptexm4xx101red-teamllm-red-teamadversarial-promptsprompt-injectionjailbreak162-transforms