Updated daily · last run Nov 20, 07:02 UTC
Trending in the bucket
Every AI dev tool the internet posted about, ranked by who's talking about it. Sorted. Mostly.
- In the bucket
- 3.1K
- Posts read
- 42.7K
- Tossed out
- 33.7K
23 tools, ranked by sources, chatter and star growth. How ranking works
- 01
#1
Live AI code reviewer that runs alongside coding agents, reviews sessions against intent, and routes work across model subscriptions.
GHHN#claude-code#codex#cursor3 posts · 6 days ago★ 136
Stars:136
- 02
#2
Desktop app that reviews GitHub pull requests with Claude Code, Codex or Gemini CLI, with findings, risk estimates and guided review.
GH#claude-code#codex#gemini-cli1 post · 14 days ago★ 14 +2/wk
Stars:14
+2 this week
- 03
#3
AI-powered Android automation tool that runs natural-language test workflows on real devices and exposes an MCP server for AI IDEs.
GHHN#mcp-server#cli#web-app4 posts · Sep 16★ 17.8K +374/wk
Stars:17.8K
+374 this week
- 04
#4
Terminal whiteboard for reviewing code diffs and agent trajectories with semantic diffs, diagrams, and pseudocode views.
Stars:3.6K
+24 this week
- 05
#5
Realtime semantic review of coding agent edits that sends rule-violation feedback back to the agent while it works.
HN#claude-code#codex#pi2 posts · Oct 13★ 150 +1/wk
Stars:150
+1 this week
- 06
#6
Visa's open-source agentic pipeline for vulnerability discovery, adversarial verification, remediation and fix validation in source code.
HN#claude#openai#any-model3 posts · Jul 25★ 3.1K +7/wk
Stars:3.1K
+7 this week
- 07
#7
Open-source CLI that runs Playwright browser tests written in plain English using AI, with reports for coding agents.
HN#cli#any-agent#macos1 post · Sep 30★ 14
Stars:14
- 08
#8
Unity Editor performance profiling agent that captures frames, generates rule-based findings with evidence chains, and offers optional LLM a
GH#any-model#mcp-server#plugin1 post · Oct 3★ 13
Stars:13
- 09#9
Desktop app that explains pull requests and codebases with spoken and written AI walkthroughs, dependency graphs, and BYOK or local models.
no public repo
- 10#10
Mac app for reviewing diffs like a pull request, with comments sent to Claude Code or Codex to fix.
HN#claude-code#codex#macos1 post · Oct 8
- 11#11
Local tool that snapshots pages before an AI agent works and pixel-diffs after, flagging changed surfaces before git commit.
HN#claude-code#cursor#copilot1 post · Aug 13
no public repo
- 12
#12
Local git proxy that validates branches with AI agents (review, tests, lint, CI repair) before pushing and opening a PR.
HN#any-agent#cli#macos1 post · Aug 22★ 8.8K
Stars:8.8K
- 13#13
AI agent that runs plain-English user tasks on iOS simulators to test if real users can complete app flows, with replayable playbooks.
no public repo
- 14#14
macOS app that shows a PR's blast radius as user journeys and runs Claude Code reviews on the change.
HN#claude-code#macos#desktop-app1 post · Sep 28
no public repo
- 15
#15
Open-source CLI that runs AI security-review agents on a repository, validates findings, and keeps results in a local SQLite ledger.
HN#any-model#cli#macos1 post · Aug 14★ 10
Stars:10
- 16
#16
Supervises multi-agent code review by running claude and codex CLI subprocesses and returning structured, auditable findings.
GH#claude-code#codex#cli1 post · Jul 26★ 99
Stars:99
- 17
#17
A single-binary web UI for reviewing a coding agent's branch like a pull request and exporting the comments as markdown for the agent to add
HN#any-agent#self-hosted#web-app1 post · Sep 21★ 14
Stars:14
- 18#18
Local-first code review tool for AI-written code with deterministic preflight, model review, and a forge daemon that never auto-merges.
HN#local-models#any-model#cli1 post · Aug 5
- 19
#19
AI code-review cockpit: Claude drafts PR reviews as local artifacts; nothing posts to GitHub until you approve.
GH#claude-code#macos#linux1 post · Aug 19★ 13
Stars:13
- 20
#20
MCP review gate that has other vendors' models review a coding agent's plan and diff, with dedup, a rounds log and a local SQLite record.
GH#mcp-server#vscode#claude-code1 post · Aug 31★ 13
Stars:13
- 21
#21
Open-source load testing tool that agents write and run; single binary, 14 protocols, CI/CD gates, built-in MCP server.
HN#claude-code#cursor#mcp-server1 post · Oct 20★ 12
Stars:12
- 22#22
Open-source Mac app that runs shell and AI review pipelines on agent branches and sends failures back to the agent until green.
HN#claude-code#codex#cursor1 post · 28 days ago
- 23
#23
Multi-agent code review for GitHub PRs using Codex and Claude Code in Rex, with a checker agent and menu bar app.
GH#claude-code#codex#macos1 post · 17 days ago★ 10
Stars:10