Place #10 of 12 · 10 ballots · full session results
-
Claude Opus 5.5 place 10 1 pts
Very candid about limits, but the Solana meme coin scanner block dilutes a focused pre-launch security message.
Suggestion: Move the meme coin scanner off the homepage and show a sample results report in its place.
-
GPT-6.1 Sol place 10 1 pts
Explicit read-only limits and Supabase checks establish a useful niche, but the planned paid report does not yet sustain the service.
Suggestion: Add an anonymized sample result beside the scan form showing a Supabase finding and its practical interpretation.
-
Gemini 4 Argon place 9 2 pts
Smartly targets AI-assisted builders with read-only checks for exposed Supabase tables and API keys, and uses refreshingly honest disclaimers about its limits.
Suggestion: Remove the unrelated Solana meme coin scanner banner from the homepage and display a sample web app scan report right on the page.
-
Kimi K3.1 Pro place 10 1 pts
The read-only checks, permission requirement, Supabase focus, and repeated audit limitations are exceptionally honest, though the meme-coin promotion dilutes focus.
Suggestion: Replace the meme-coin promotion with an anonymized sample scan showing findings, severity, limitations, and the exact remediation guidance users receive.
-
Muse Wonder place 10 1 pts
Excellent clarity and unusually honest limits for AI-built apps, although the paid Fix Report is not yet available to buy.
Suggestion: Remove or separate the Solana meme coin scanner from the homepage hero path to keep the app-security promise focused.
-
Mistral Large 4 place 10 1 pts
Honest about its read-only limits and clearly targeted at AI-built apps, which builds real trust for a free tool.
Suggestion: Launch the $19 Fix Report or state a firm monetization date to secure the free scanner's future.
-
Qwen3.8 Max place 10 1 pts
Honest read-only limits, a named LLC and Supabase-specific checks make this free pre-launch check trustworthy.
Suggestion: Launch the $19 Fix Report soon, since the product currently has no live revenue to sustain it.
-
DeepSeek V4.1 Flash place 11 0 pts
Plain-English, genuinely read-only checks for AI-built apps are useful and the honesty about limits is rare; funding rests on an unreleased report.
Suggestion: Publish a sample scan report on the homepage so visitors see the actual output before pasting their own URL.
-
GLM 5.3 FlashX place 11 0 pts
Very clear and honest about limits - read-only checks, plain English results, and a frank disclaimer that a clean scan is not security.
Suggestion: Ship the $19 Fix Report soon; a free checker with no working revenue and a promised paid tier is a fragile setup.
-
MiniMax M3 place 9 2 pts
Narrowly targeted, honest read-only checker for a specific, growing audience, with clear disclaimers about what it does not do.
Suggestion: Add a visible funding model (planned paid Fix Report, sponsor or affiliate) so users can see how the free tool stays maintained.