skills.rest
JEROME-PRAKASH-L
JEROME-PRAKASH-LCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

openclaw-qa-testing

Run, debug, and extend OpenClaw qa-lab and qa-channel end-to-end test scenarios.

1|Updated Jun 9, 2026
Search Tags:#test automation#credentials#e2e testing#openclaw#qa testing#live lanes#character eval
One-click install
npx skills add https://github.com/JEROME-PRAKASH-L/openclaw --skill openclaw-qa-testing-jerome-prakash-l
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: openclaw-qa-testing
Source: https://github.com/JEROME-PRAKASH-L/openclaw/tree/main/.agents/skills/openclaw-qa-testing
Command: npx skills add https://github.com/JEROME-PRAKASH-L/openclaw --skill openclaw-qa-testing-jerome-prakash-l

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Running and debugging OpenClaw's repo-local QA infrastructure (qa-lab suites, qa-channel scenarios, live model lanes, and credential-backed channel tests) requires knowing many scattered commands, model policies, and credential sources, which this Skill consolidates into one operational guide. ## Core Features & Use Cases - QA Suite Execution: Run mock or live-frontier qa-lab suites with the correct model policy (openai/gpt-5.4 in fast mode), watch the live UI port, and collect summary and report artifacts under .artifacts/qa-e2e/. - Live Channel & Credential Management: Handle Telegram, WhatsApp, Matrix, and Convex-backed QA credentials via 1Password or Convex leasing, including the npm Telegram Docker lane and GitHub workflow dispatch fallbacks. - Character Evals & Model Lanes: Run multi-model persona evaluations with judge models, plus Codex CLI model lanes and OTEL trace smoke tests. - Use Case: A maintainer needs to validate a release candidate; use this Skill to run the full live-frontier suite, dispatch the NPM Telegram Beta E2E workflow, and report pass/fail counts with artifact paths. ## Quick Start Use the openclaw-qa-testing skill to run the full qa-lab suite in live-frontier mode with openai/gpt-5.4 and report the artifact paths and pass/fail counts.

Frequently Asked Questions about openclaw-qa-testing

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run the OpenClaw qa-lab suite against a live model?▼

Run pnpm openclaw qa suite with --provider-mode live-frontier and --model openai/gpt-5.4, passing OPENCLAW_LIVE_OPENAI_KEY from your OpenAI API key. Results are written to a summary JSON and markdown report under .artifacts/qa-e2e/.

How do I run character evals across multiple models?▼

Use pnpm openclaw qa character-eval with --model entries shaped as provider/model,thinking=<level> and separate --judge-model flags. The report includes judge rankings, run stats, durations, and full transcripts.

Where are OpenClaw QA Telegram credentials stored?▼

Telegram live test secrets live in the 1Password vault OpenClaw under the item Telegram E2E, accessed with op inside tmux. Alternatively, Convex credential leasing provides pooled Telegram tokens and group ids for shared QA infrastructure.

Which OpenAI models are allowed for live QA lanes?▼

The live OpenAI lane uses openai/gpt-5.4 with fast mode enabled. The gpt-5.4-pro and gpt-5.4-mini variants are explicitly disallowed unless the user explicitly requests a model policy change.

Why does the Codex CLI model lane return fallback or auth text every turn?▼

This usually means CODEX_HOME is not preserved or auth config is missing in the sandboxed gateway child process. Check CODEX_HOME, secret-backed auth, and gateway child logs before changing scenario assertions.

What should I do when a qa-lab scenario fails?▼

Fix the product or harness root cause rather than the assertion, then rerun the full lane. Common causes include too-short live timeouts, discovery prompts not pointing at repo/ paths, and brittle subagent proof checks.

Related Skills

Explore More Software Engineering Skills→
openclawopenclaw
388.4k

openclaw-qa-testing

Run, debug, and extend OpenClaw qa-lab and qa-channel test scenarios.

Official
Advanced
glowingkittyglowingkitty
43

fix-next-test

Fixes the next prioritized failing Playwright regression test after triage and verification.

Community
Advanced
AI-Foundry-CoreAI-Foundry-Core
4

python-observability

Configure structured logging, metrics, and tracing for Python applications with OpenTelemetry.

Official
Advanced
programadorisgodprogramadorisgod
1

frontend-testing-best-practices

Provides E2E-first frontend testing guidelines for React, Next.js, and Svelte projects.

Community
Advanced