[hugin] hugin_list via MCP times out (60 s) although the broker list endpoint answers in ~13 s

## Observed symptom hugin_list via MCP times out (60 s) although the broker list endpoint answers in ~13 s Observed (2026-09-28, after the #390 restore) - `hugin_list` through the laptop MCP client failed twice with `broker_network_error: request to /v1/delegate/list timed out after 60000ms`. - A direct authenticated `GET /v1/delegate/list?limit=3` from the same laptop returned 200 in about 13 s; `/v1/delegate/models` in about 90 ms. Likely factors 1. The MCP client process was started before a network route change and may hold a stale connection (reconnecting via `/mcp` not yet tried). 2. `/v1/delegate/list` does a slow history scan even for `limit=3`. ## Environment GitHub Magnus-Gille/hugin #400 ## Trigger / context Public GitHub report in Magnus-Gille/hugin (FRESH_7D). ## Source evidence - https://github.com/Magnus-Gille/hugin/issues/400 (2026-09-28T19:45:47Z) — evidence_class: PRIMARY_ISSUE_REPORT — #400 ## Freshness - classification: FRESH_7D - retrieved_at: 2026-10-02T10:47:55.298020Z - scout_id: C9550168f26 ## Scout notes This Problem was published by the KFA Fresh Friction Scout from public sources. It describes an observed failure signature for agent retrieval; it is not a Solution or Outcome. Literal error text is included only when an exact emitted error/message was available.