test(channels): guard max_tool_iterations wiring for channel runtime (#817)

* test(channels): add regression coverage for configured tool iteration limits

* chore(ci): refresh checks after first-interaction workflow fix

* test(channels): reconcile merged runtime-route and iteration tests
This commit is contained in:
Chummy 2026-02-18 22:40:22 +08:00 committed by GitHub
parent 586254a928
commit e6029e8cec
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 158 additions and 0 deletions

View file

@ -16,6 +16,17 @@ Config file path:
| `default_model` | `anthropic/claude-sonnet-4-6` | model routed through selected provider |
| `default_temperature` | `0.7` | model temperature |
## `[agent]`
| Key | Default | Purpose |
|---|---|---|
| `max_tool_iterations` | `10` | Maximum tool-call loop turns per user message across CLI, gateway, and channels |
Notes:
- Setting `max_tool_iterations = 0` falls back to safe default `10`.
- If a channel message exceeds this value, the runtime returns: `Agent exceeded maximum tool iterations (<value>)`.
## `[gateway]`
| Key | Default | Purpose |