AI SummaryQA is a skill booster that enables developers to quickly create and manage stress test conversations with AI-generated members in Convos, streamlining QA testing workflows for messaging systems and invite flows.
Install
Copy this and paste it into Claude Code, Cursor, or any AI assistant:
I want to install the "QA" skill in my project. Please run this command in my terminal: # Install skill into the correct directory (7 files) mkdir -p .claude/skills/commands && curl --retry 3 --retry-delay 2 --retry-all-errors -o .claude/skills/commands/SKILL.md "https://raw.githubusercontent.com/xmtplabs/convos-ios/dev/.claude/commands/qa.md" && curl --retry 3 --retry-delay 2 --retry-all-errors -o .claude/skills/commands/build.md "https://raw.githubusercontent.com/xmtplabs/convos-ios/dev/.claude/commands/build.md" && curl --retry 3 --retry-delay 2 --retry-all-errors -o .claude/skills/commands/firebase-token.md "https://raw.githubusercontent.com/xmtplabs/convos-ios/dev/.claude/commands/firebase-token.md" && curl --retry 3 --retry-delay 2 --retry-all-errors -o .claude/skills/commands/format.md "https://raw.githubusercontent.com/xmtplabs/convos-ios/dev/.claude/commands/format.md" && curl --retry 3 --retry-delay 2 --retry-all-errors -o .claude/skills/commands/lint.md "https://raw.githubusercontent.com/xmtplabs/convos-ios/dev/.claude/commands/lint.md" && curl --retry 3 --retry-delay 2 --retry-all-errors -o .claude/skills/commands/setup.md "https://raw.githubusercontent.com/xmtplabs/convos-ios/dev/.claude/commands/setup.md" && curl --retry 3 --retry-delay 2 --retry-all-errors -o .claude/skills/commands/test.md "https://raw.githubusercontent.com/xmtplabs/convos-ios/dev/.claude/commands/test.md" Then restart Claude Code (or reload the window in Cursor) so the skill is picked up.
Description
Create and manage Convos QA stress test conversations
Usage
Use natural language. Examples: ` /qa create a convo with 5 members /qa make a conversation about travel with 10 members and join it in the simulator /qa join in the simulator a convo with 8 members /qa 5 members talking about coffee, join in sim /qa add 3 fake members to https://convos.app/v2?i=abc123 /qa what's running? /qa stop stress-test-abc123 /qa restart in prod /qa switch to dev /qa update `
QA
Create and manage stress test conversations for Convos QA testing.
When to Use
• Creating test conversations with fake AI-generated members • Stress testing the messaging system • Testing invite flows in the iOS simulator • Adding fake members to existing conversations
Step 0: Bootstrap (Run Every Time)
Before executing any QA command, ensure the convos-agents service is running. 0.1: Check if Service is Already Running `bash curl -s http://localhost:3000/health 2>/dev/null | grep -q '"status":"ok"' ` If health check passes, skip to Step 1 (Execute Command). If service is NOT running, continue with bootstrap steps 0.2-0.5. 0.2: Check GitHub CLI Authentication `bash gh auth status ` If not authenticated: ` ❌ Not logged into GitHub CLI. Run: gh auth login ` Stop and wait for user to authenticate. 0.3: Clone or Update the Repository Check if the repo exists: `bash ls ~/.convos-qa/convos-ai/package.json 2>/dev/null ` If repo doesn't exist, clone it: `bash mkdir -p ~/.convos-qa gh repo clone xmtplabs/convos-ai ~/.convos-qa/convos-ai ` If repo exists, pull latest: `bash cd ~/.convos-qa/convos-ai && git pull ` 0.4: Check for OPENAI_API_KEY Check if ~/.convos-qa/.env exists and contains OPENAI_API_KEY: `bash grep -s "^OPENAI_API_KEY=" ~/.convos-qa/.env | cut -d'=' -f2 ` If empty or file doesn't exist, ask the user: ` 🔑 OpenAI API key required for AI-generated conversations. Enter your OpenAI API key (starts with sk-): ` Save the key: `bash mkdir -p ~/.convos-qa echo "OPENAI_API_KEY=<user-provided-key>" > ~/.convos-qa/.env ` 0.5: Install Dependencies, Build, and Start the Service Install dependencies, build TypeScript, and start the service: `bash cd ~/.convos-qa/convos-ai
Discussion
Health Signals
My Fox Den
Community Rating
Sign in to rate this booster