Menu
Test the generated project inside an E2B cloud sandbox via Claude Code, apply fixes iteratively, and produce a quality report
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Search and install agent skills from ClawHub, the public skill registry.
Schedule reminders and recurring tasks.
Remote-control tmux sessions for interactive CLIs by sending keystrokes and scraping pane output.
Instinct-based learning system that observes sessions via hooks, creates atomic instincts with confidence scoring, and evolves them into skills/commands/agents. v2.1 adds project-scoped instincts to prevent cross-project contamination.
Research a hackathon using a user-supplied URL plus optional doc URLs (sitemap-driven scrape).
Two-layer memory system with grep-based recall.
| name | tester |
| description | Test the generated project inside an E2B cloud sandbox via Claude Code, apply fixes iteratively, and produce a quality report |
| metadata | {"openclaw":{"always":false}} |
Validate the generated project inside an E2B cloud sandbox: Claude Code installs dependencies, runs tests, applies fixes iteratively, and writes a structured report. The entire phase runs in an isolated Linux VM with full internet access.
After coder agents have finished implementing. Requires hackathon/project/ to exist.
Phase 6 uses E2BTestingExecutor which:
hackathon/project/ and sibling artifacts (plan.md, tasks.json, etc.).claude --dangerously-skip-permissions inside the sandbox with the
testing prompt — CC has full autonomy to install packages, run tests, fix
issues, and start servers.test_results.json from the sandbox to the local workspace.The sandbox provides full isolation from the host system with internet access, so pip/npm install and server startup work without restrictions.
hackathon/test_results.json