ワンクリックで
evor-dashboard
Start the live FastAPI + SSE dashboard for the active Evor mission on port 8756
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Start the live FastAPI + SSE dashboard for the active Evor mission on port 8756
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Mission interview workflow that elicits all mission settings and launches a validated run via evor_init_run
Scan an existing ML workspace to produce a starting-point report; pre-fills the evor-setup interview
Check environment health and .evor integrity; auto-repair obvious issues
Inspect accumulated Gotchas (failures + hardware limits) and the hardware capability profile
Evor MCP tool catalog, run lifecycle, artifact schemas, and native-tool guide. Auto-loads when operating an evor run — setting up/running/resuming/inspecting a mission, recording nodes/evals, launching or watching training, reading/writing tick artifacts, citing papers, emitting/querying signals, or changing any evor state.
Generate a final Evor mission report with tree visualization, frontier table, and static HTML export
| name | evor-dashboard |
| description | Start the live FastAPI + SSE dashboard for the active Evor mission on port 8756 |
| argument-hint | [run-id] |
| level | 2 |
| skills | ["oh-my-evor:evor-mcp"] |
<Use_When>
/evor-dashboard<Do_Not_Use_When>
evor-report.evor/active-run.json absent) — redirect to /evor-setup
</Do_Not_Use_When>Call evor_state_read to get run_dir from the active run. If a run-id argument was provided, call evor_state_read with the provided run-id to get run_dir.
If no active run is found: print "No active run found. Start a mission with /evor-setup first." and stop.
lsof -i :8756 2>/dev/null | grep LISTEN
If a server is already listening on port 8756: print "Dashboard already running at http://localhost:8756 (run-id: <run_id>)" and open the browser (Step 4). Do not start a second server.
Launch the server in the background. Call evor_state_read to get the log path from the active run state, then:
EVOR_RUN_DIR=<run_dir> uvicorn evor.dashboard.server:app \
--host 0.0.0.0 \
--port 8756 \
--reload \
--reload-dir <run_dir> \
2><log_path> &
Wait up to 5 seconds for the server to become ready:
for i in 1 2 3 4 5; do
curl -s http://localhost:8756/health && break
sleep 1
done
If health check fails after 5 attempts: print "Dashboard failed to start. Check the server log for details." and stop.
Attempt to open the dashboard in the default browser:
xdg-open http://localhost:8756 2>/dev/null || open http://localhost:8756 2>/dev/null || true
Print:
Dashboard running at: http://localhost:8756
Run ID: <run_id>
Mission: <mission_id>
Available views:
/ — Evolution tree (D3 DAG, frontier highlighted)
/telemetry — Live training curves (Chart.js, SSE stream)
/frontier — Frontier table (best-so-far nodes)
/health — Server health check
Press Ctrl+C or run /evor-dashboard --stop to shut down.
If the user passes --stop as an argument:
pkill -f "uvicorn evor.dashboard.server:app"
Print: "Dashboard server stopped."
<Tool_Usage>
evor_state_read — read active-run.json to resolve run_dirBash — lsof, uvicorn background launch, curl health check, xdg-open/open
</Tool_Usage>