메뉴
Run the full Explorer end-to-end wheel test against a real LLM provider and Memgraph. Use to verify the Explorer produces valid wheels after changes to the exploration pipeline, transformations, or synthesis.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | df-wheel-test |
| description | Run the full Explorer end-to-end wheel test against a real LLM provider and Memgraph. Use to verify the Explorer produces valid wheels after changes to the exploration pipeline, transformations, or synthesis. |
| disable-model-invocation | true |
| allowed-tools | Bash(poetry run pytest *) |
Usage: /df-wheel-test
Runs the Explorer wheel end-to-end test with a real LLM provider. Requires Memgraph running (docker compose -f docker-compose.test.yml up -d) and provider credentials in .env.
poetry run pytest "tests/test_agents_e2e.py::TestExplorerEndToEnd::test_explorer_produces_wheels" --real-llm -v -s --tb=short
Review the layered prompt system that assembles the framework's dialectical reasoning — across apps, agents, concerns, and shared theory. Reviews at three altitudes — the isolated prompt, the assembled context it lands in, and the whole reasoning chain it steers. Use proactively when writing or modifying LLM prompts in agents/, concerns/, or orchestrator/tools/.
Sync the theory→implementation wiki (docs/theory/) with the Structured Dialectics theory PDFs and the codebase. Ingests/validates theory changes, updates the wiki's claim→code mapping and statuses, promotes essentials to CLAUDE.md, and reports theory↔implementation gaps. Use after theory PDFs change, after implementing/changing theory-encoding code, or to audit drift.
Manage the test Memgraph container (start/stop/restart/status/logs/clear/wipe) via docker-compose.test.yml. Use when graph tests need the DB up, data cleared, a stuck lock cleared, or the volume fully wiped.