New in the bucket
Tools by when we first saw them, newest first.
115 tools
- 01
#1
Skill and CLI for coding agents that generates browser review reports from a branch, with tests, UI screenshots and database change views.
HN#claude-code#codex#cursor1 post · 2 days ago★ 115
Stars:115
- 02
#2
Pytest plugin that records test dependencies and runs only the tests a change can affect, for AI coding agents and CI.
GH#claude-code#codex#cursor1 post · 3 days ago★ 12
Stars:12
- 03#3
AI code reviewer that analyzes a repository's conventions and reviews every pull request against them.
HN#github-action1 post · 3 days ago
no public repo
- 04
#4
Claude Code plugin running parallel AI testers on iOS simulators and Android emulators via GitHub Actions, producing bug reports with screen
GH#claude-code#plugin#github-action1 post · 4 days ago★ 47
Stars:47
- 05
#5
CLI that captures coding agent corrections as ast-grep, Vale, or runtime rules enforced across editors, agents, and CI.
HN#claude-code#cursor#opencode1 post · 5 days ago★ 118
Stars:118
- 06#6
Open-source multi-agent system that runs the full QA test process from requirements review to defect reporting, self-hosted with any model.
HN#self-hosted#any-model#web-app1 post · 7 days ago
- 07
#7
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 · 13 days ago★ 14 +2/wk
Stars:14
+2 this week
- 08
#8
Headless browser QA tool that crawls apps, traps runtime/network/layout/a11y failures, and returns bug reports for coding agents to fix.
GH#claude-code#codex#cursor1 post · 13 days ago★ 48 +35/wk
Stars:48
+35 this week
- 09#9
Independent verifier for C/C++ and firmware bug fixes that reproduces bugs, checks fixes, and detects masking, with no LLM in the verdict.
HN#claude-code#codex#mcp-server2 posts · 9 days ago
no public repo
- 10
#10
Open-source AI code review agent that runs /review, /improve and /describe commands on pull requests across major git providers.
GH#github-action#cli#any-model1 post · 15 days ago★ 16.5K +419/wk
Stars:16.5K
+419 this week
- 11
#11
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 · 16 days ago★ 10
Stars:10
- 12
#12
Offline diff reviewer that catches weakened tests, loosened assertions and lowered coverage, and sends coding agents back to fix the code.
HN#claude-code#codex#cursor2 posts · 8 days ago★ 11
Stars:11
- 13#13
AI QA agent that authors, runs, and auto-heals end-to-end tests from plain-English flows, with a CLI for CI and MCP integration for coding a
HN#any-agent#cli#web-app1 post · 18 days ago
no public repo
- 14
#14
CLI and MCP server that visualizes Vivado/Zynq/PYNQ projects and checks clock-domain crossings, constraints and build failures for AI agents
GH#claude-code#codex#gemini-cli1 post · 25 days ago★ 28
Stars:28
- 15
#15
Lints rendered UI in a real browser for overflow, contrast, overlap and layout defects, built as a quality gate for AI coding agents.
GH#claude-code#cursor#codex1 post · 26 days ago★ 686 +258/wk
Stars:686
+258 this week
- 16#16
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 · 27 days ago
- 17
#17
Open-source E2E testing framework for web and mobile apps with AI agent steps, provider emulators, fault injection, and an MCP server.
HN#claude-code#cursor#codex1 post · 28 days ago★ 18
Stars:18
- 18
#18
CLI that reviews coding agent sessions and diffs, checking whether asks were done, claims hold, and the code is correct.
HN#claude-code#codex#cursor1 post · 28 days ago★ 10
Stars:10
- 19
#19
Codex skill that breaks AI-written code into explained functional blocks and verifies goals against recorded evidence, generating HTML/Markd
Stars:134
+1 this week
- 20#20
GitHub App that reviews pull requests with Gemini, posting line-by-line comments and critical issue flags.
HN#gemini#github-action#web-app1 post · Oct 20
no public repo
- 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
GitHub app that uses static analysis and AI to analyze, fix and merge Dependabot dependency-update pull requests.
HN#github-action#any-model1 post · Oct 20
no public repo
- 23
#23
SDK and MCP server that lets coding agents verify their changes against the running app, returning pass/fail verdicts with file:line evidenc
HN#claude-code#cursor#codex1 post · Oct 20★ 1.3K +40/wk
Stars:1.3K
+40 this week
- 24
#24
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 · 5 days ago★ 136
Stars:136
- 25#25
A workflow engine for agentic development that hard-bonds agent claims to deterministic checks run by the tool, not the LLM.
HN#claude-code#codex#pi1 post · Oct 12
no public repo