AI coding tools / Alternatives

Cursor Marketplace security notice (Claude Code extension) alternatives, tested on day 7

The other AI coding tools Day 7 has checked, ranked by stamp: Delivered first, then Partly, Not yet and Too early. Within a stamp, more user reports rank higher.

Cursor Marketplace security notice (Claude Code extension) itself is stamped Too early: read its receipt.

1. 01

   No. 003 · OpenAI · launched 25 Sept 2026 · stamped 30 Oct 2026

   Codex CLI 0.157.0

   Claims ✓0 ~1 ✗0 ?3 · 8 reports · vs Cursor Marketplace security notice (Claude Code extension)

   The release shipped with its background server, new models and fixes. Some Windows and Arch Linux users hit startup problems with the new server.

   Delivered
2. 02

   No. 006 · Anomaly (OpenCode) · launched 14 Oct 2026 · stamped 30 Oct 2026

   OpenCode 2.0

   Claims ✓2 ~1 ✗1 ?1 · 17 reports · vs Cursor Marketplace security notice (Claude Code extension)

   The background service and Code Mode shipped as promised, but "backwards compatible" is wrong: V1 plugins and the old server API break.

   Partly
3. 03

   No. 020 · Anthropic · launched 30 Oct 2026 · stamped 31 Oct 2026

   Claude Ultra (no 5-hour limit)

   Claims ✓2 ~0 ✗0 ?3 · 4 reports · vs Cursor Marketplace security notice (Claude Code extension)

   Day 1: one hands-on account confirms the five-hour bar is gone; users who say they have paid for months still have it. Rollout is staged.

   Too early
4. 04

   No. 016 · OpenAI · launched 29 Oct 2026 · stamped 31 Oct 2026

   ChatGPT Pro: 2x Work and Codex usage

   Claims ✓1 ~1 ✗0 ?3 · 4 reports · vs Cursor Marketplace security notice (Claude Code extension)

   Day 2: users on the lowest Pro tier saw their meters double the same day, but the 'doubling all usage limits' headline is wider than OpenAI's own release notes. No figures published.

   Too early
5. 05

   No. 015 · GitHub · launched 28 Oct 2026 · stamped 30 Oct 2026

   Copilot Fusion

   Claims ✓2 ~0 ✗1 ?0 · 4 reports · vs Cursor Marketplace security notice (Claude Code extension)

   Generally available and the benchmark checks out, but "all plans in all four clients" does not: top tiers are paid-only and two clients have gaps.

   Too early
6. 06

   No. 018 · Factory · launched 29 Oct 2026 · stamped 31 Oct 2026

   Droid 1.0

   Claims ✓0 ~0 ✗0 ?3 · 1 reports

   Launched two days ago; the speed and accuracy numbers are Factory's own and no independent reproduction or user reports yet.

   Too early

Ranking uses only the stamp and report count on each receipt. Nothing here is paid placement. How we stamp · All AI coding tools · Submit a launch