| name | investigate |
| description | Debug Ethereum devnet or network issues. Use when diagnosing finality delays, network splits, offline nodes, client bugs, or general network health problems. Works for both local Kurtosis devnets and live public devnets. |
| argument-hint | <network-name and/or issue description> |
| user-invocable | false |
Investigate Ethereum Network Issues
This skill resolves the target network, then hands off to the debugging runbook. One
runbook owns the procedure for every network kind — only the access layer differs.
The user MUST specify which network to debug. If not provided, ask them.
Step 1: Resolve the network target
Run both checks in parallel:
kurtosis enclave ls 2>/dev/null
panda datasources --json 2>/dev/null
- Target matches a Kurtosis enclave name →
kind: local.
- Target is found in panda datasources (Dora networks, or container logs in
external.otel_logs on clickhouse-raw) → kind: public.
- Found in neither → stop and tell the user the network was not found in any local
enclave or remote datasource.
Step 2: Follow the debugging runbook
panda read runbooks://debug_ethereum_network
Follow it with the resolved target and the user's symptom. It owns the symptom→branch
table, the CL-vs-EL localization matrix, and the per-kind access table (local OTel via
local-kurtosis vs public-devnet external.otel_logs, Kurtosis ports vs published
endpoints). Unsure which runbook applies at any point →
panda search runbooks "<symptom>".
Notes
- The runbooks are the source of truth for the debugging path — this skill only
resolves the target and routes.
- Use
panda search examples for query patterns before writing complex queries from
scratch.
- Produce a debug report file at
/workspace/ and give the user the path at the end.