Explore / GitHub Agent Apps Campaign

GitHub Agent Apps Campaign

DeepQA runs its QA engine against agent apps that are trending on GitHub: workflow builders, chat UIs, coding agents, research agents and observability tools. Open-source apps are cloned and booted from their repository, hosted apps are tested in place, and every report is published here.

A team that wants its report taken down can ask us and we will remove it.

56
apps
56
public reports
631
scenarios
562
scenarios passed
28
issues confirmed
54
re-verified live
2k
screenshots
9.6k
browser actions
9.1k
model calls
57.2M
tokens
8.5 h
testing time

56 apps (showing 13 to 24)

demo.sourcebot.dev in the browser during the run
Sourcebot demodemo.sourcebot.devOpen app ↗sourcebot-dev/sourcebotHosted appAI code searchHosted, in placerun by DeepQA Team

Public demo of Sourcebot, an open-source code search and Ask AI tool across repositories: search, repository browsing, scoped chat and model choice. Tested in place without signing in.

Run #1succeeded12 scenarios, 8 pass, 2 fail, 2 blocked
1 issue1
www.aishort.top in the browser during the run
AiShortwww.aishort.topOpen app ↗rockbenben/ChatGPT-ShortcutHosted appPrompt libraryHosted, in placerun by DeepQA Team

Open-source prompt library with tags, search, favorites and community prompts for ChatGPT, Claude, Gemini and DeepSeek. Tested in place on its hosted instance.

Run #1succeeded12 scenarios, 11 pass, 1 fail
0 issues
microsoft-omniparser-v2.hf.space in the browser during the run
OmniParser V2 demomicrosoft-omniparser-v2.hf.spaceOpen app ↗microsoft/OmniParserHosted appGUI agent screen parserHosted, in placerun by DeepQA Team

Microsoft's public Hugging Face Space of OmniParser V2, an open-source screen parsing tool for vision based GUI agents. Tested in place.

Run #1succeeded9 scenarios, 7 pass, 1 fail, 1 blocked
1 issue1
sickn33.github.io in the browser during the run
Agentic Awesome Skills workbenchsickn33.github.ioOpen app ↗sickn33/agentic-awesome-skillsHosted appAgent skills workbenchHosted, in placerun by DeepQA Team

Public workbench of Agentic Awesome Skills, an open-source catalog of agent skills: load example stacks, preview installs, import plans and shortlist skills. Tested in place.

Run #1succeeded12 scenarios, 11 pass, 0 fail, 1 blocked
0 issues
ChatAny in the browser during the run
ChatAnyChatAnyChatAnyTeam/ChatAnyChat UISandbox, from sourcerun by DeepQA Team

Open-source multi-model chat UI with custom masks, templates and conversation history. Run from its repository in the DeepQA sandbox.

Run #1succeeded12 scenarios, 8 pass, 2 fail, 2 blocked
0 issues
TranslateBooksWithLLMs in the browser during the run
TranslateBooksWithLLMsTranslateBooksWithLLMshydropix/TranslateBooksWithLLMsLLM translation appSandbox, from sourcerun by DeepQA Team

Open-source web app that translates books, subtitles and documents with local or hosted LLMs, with glossaries, styles, checkpoints and settings. Run from its repository in the DeepQA sandbox.

Run #1succeeded12 scenarios, 10 pass, 2 fail
0 issues
aisheets in the browser during the run
AI Sheetsaisheetshuggingface/aisheetsAI dataset spreadsheetSandbox, from sourcerun by DeepQA Team

Hugging Face's open-source tool to build, enrich and transform datasets with AI models in a spreadsheet UI. Run from its repository in the DeepQA sandbox.

Run #1succeeded12 scenarios, 11 pass, 1 fail
0 issues
mcphub in the browser during the run
MCPHubmcphubsamanhappy/mcphubMCP server hubSandbox, from sourcerun by DeepQA Team

Open-source hub to manage and route Model Context Protocol servers from one dashboard: servers, resources, users, logs and settings. Run from its repository in the DeepQA sandbox.

Run #1succeeded12 scenarios, 11 pass, 0 fail, 1 blocked
0 issues
Claudable in the browser during the run
ClaudableClaudableanymorph-ai/ClaudableAI app builderSandbox, from sourcerun by DeepQA Team

Open-source app builder that drives coding agents from a web workspace with preview, code view, agent and model choice. Run from its repository in the DeepQA sandbox.

Run #1succeeded12 scenarios, 10 pass, 0 fail, 2 blocked
0 issues
all-model-chat.pages.dev in the browser during the run
AMC WebUIall-model-chat.pages.devOpen app ↗yeahhe365/AMC-WebUIHosted appChat UIHosted, in placerun by DeepQA Team

Open-source bring-your-own-key chat UI for multiple model providers with presets, a library and live artifacts. Tested in place on its hosted instance.

Run #1succeeded10 scenarios, 10 pass, 0 fail
0 issues
tiktokenizer.vercel.app in the browser during the run
Tiktokenizertiktokenizer.vercel.appOpen app ↗dqbd/tiktokenizerHosted appTokenizer playgroundHosted, in placerun by DeepQA Team

Open-source playground that shows how chat messages and raw text are tokenized for different LLMs. Tested in place on its hosted instance.

Run #1succeeded12 scenarios, 12 pass, 0 fail
0 issues
promptfoo in the browser during the run
promptfoopromptfoopromptfoo/promptfooLLM evals and red teamingSandbox, from sourcerun by DeepQA Team

Open-source tool for testing, evaluating and red teaming LLM apps, with a web UI for eval setup, red team wizard, model audit and settings. Run from its repository in the DeepQA sandbox.

Run #1succeeded12 scenarios, 11 pass, 0 fail, 1 blocked
0 issues

Put an agent team on your next pull request.

Connect a repo, dispatch a Run, and read an audited, evidence-backed report the same day.