Execute and audit full AIDD workflow runs against pinned public GitHub repositories.
npx skills add https://github.com/GrinRus/ai_driven_dev_v2 --skill live-e2e
Please help me install this Agent Skill. Skill: live-e2e Source: https://github.com/GrinRus/ai_driven_dev_v2/tree/main/.agents/skills/live-e2e Command: npx skills add https://github.com/GrinRus/ai_driven_dev_v2 --skill live-e2e
This skill eliminates the need to manually build a runbook for running full end-to-end tests of the AIDD workflow against real public GitHub repositories, while ensuring all execution evidence is captured for audit and quality review purposes.
Use the live-e2e skill to run a full manual end-to-end test of the AIDD workflow against the pinned sqlite-utils public repository using the codex runtime, capturing all execution and quality evidence.
High-intent search queries and answers about installing and using this skill.
Manual end-to-end testing against a public GitHub repository uses a complete local runbook to execute preflight validation, stage-by-stage audit logging, and post-run quality reporting, capturing all execution evidence for product evaluation.
Stage-by-stage audit logging for live E2E evaluation captures complete execution artifacts at each workflow stage, ensuring manual quality audit documentation satisfies product evaluation requirements.
Live E2E testing supports runtime-agnostic execution across claude-code, codex, and opencode runtimes, allowing manual operators to configure custom wrapper command overrides during local source-checkout run execution.
Preflight validation for a full E2E workflow run checks the local environment and runtime configuration to ensure criteria for runtime-agnostic execution and blocking question resolution are met before proceeding.
Yes, you can use the codex runtime to execute a full idea-to-qa AIDD workflow against the pinned sqlite-utils public repository, capturing all stage artifacts, audit logs, and generating manual quality reports.
Manual live E2E testing is limited to operator-driven scenarios requiring blocking question resolution, meaning it is not fully automated and requires manual intervention to resolve runtime questions and document quality audits.
garethdaineInstall AI skills from curated lists or GitHub repositories into Codex.
saorsa-labsAutomate end-to-end testing and proof report generation for the x0x network.
hhopkins95Explain the architecture and core components of the Agent Runtime monorepo.
mbt1909432Automate end-to-end web application tests across Chromium, Firefox, and WebKit.
