fix(display): show reasoning before response when tool calls suppress content by teknium1 · Pull Request #3566 · NousResearch/hermes-agent

teknium1 · 2026-03-28T19:32:29Z

Problem

When reasoning display is enabled (/reasoning show) and the model generates content with <REASONING_SCRATCHPAD> tags alongside tool calls, the reasoning box appeared AFTER the already-streamed response — reversed from the expected order.

Root Cause

In the chat completions streaming loop (run_agent.py line 3902), content deltas are suppressed from the stream callback once any tool call delta arrives (if not tool_calls_acc). This is intentional — it prevents chatty "I'll use the tool..." text from displaying alongside tool execution.

But it also suppresses reasoning tags in the content from reaching the CLI's _stream_delta tag extraction. Since the reasoning is never extracted during streaming, _reasoning_shown_this_turn stays False, and the post-response fallback fires — displaying the reasoning BELOW the already-visible streamed response.

Fix

Route suppressed content to stream_delta_callback even when tool calls are present. The CLI's _stream_delta handles tag extraction — reasoning tags are routed to the reasoning display box, while non-reasoning text is handled by the existing stream display logic.

Before/After

Before: Reasoning → Tool calls → Response → Reasoning box (wrong — reasoning shown after response)
After: Reasoning box → Tool calls → Response (correct order)

… user override The minimax-specific auto-correction in runtime_provider.py was preventing users from overriding to the OpenAI-compatible endpoint via MINIMAX_BASE_URL. Users in certain regions get nginx 404 on api.minimax.io/anthropic and need to switch to api.minimax.chat/v1. The generic URL-suffix detection already handles /anthropic → anthropic_messages, so the minimax-specific code was redundant for the default path and harmful for the override path. Now: default /anthropic URL works via generic detection, user override to /v1 gets chat_completions mode naturally. Closes #3546 (different approach — respects user overrides instead of changing the default endpoint).

…ppress content When a model generates content (containing <REASONING_SCRATCHPAD> tags) alongside tool calls in the same API response, content deltas were suppressed from streaming once any tool call chunk arrived. This prevented the CLI's tag extraction from running, so reasoning was never shown during streaming. The post-response fallback then displayed reasoning AFTER the already-visible streamed response, creating a confusing reversed order. Fix: route suppressed content to stream_delta_callback even when tool calls are present. The CLI's _stream_delta handles tag extraction — reasoning tags are routed to the reasoning display box, while non-reasoning text is handled by the existing stream display logic. This ensures reasoning appears before tool execution and the final response, matching the expected visual order.

github-actions · 2026-03-28T19:32:40Z

⚠️ Supply Chain Risk Detected

This PR contains patterns commonly associated with supply chain attacks. This does not mean the PR is malicious — but these patterns require careful human review before merging.

⚠️ WARNING: Install hook files modified

These files can execute code during package installation or interpreter startup.

Files:

hermes_cli/setup.py

Automated scan triggered by supply-chain-audit. If this is a false positive, a maintainer can approve after manual review.

PR #3566 intentionally routes suppressed content to stream_delta_callback when tool calls are present, so reasoning tag extraction can fire during streaming. The test was still asserting the old behavior where content after tool calls was fully suppressed from the callback. Updated the assertion to match: content IS delivered to the callback (for tag extraction), with display-level suppression handled by the CLI's _stream_delta.

…3574) PR #3566 intentionally routes suppressed content to stream_delta_callback when tool calls are present, so reasoning tag extraction can fire during streaming. The test was still asserting the old behavior where content after tool calls was fully suppressed from the callback. Updated the assertion to match: content IS delivered to the callback (for tag extraction), with display-level suppression handled by the CLI's _stream_delta.

… content (NousResearch#3566) * fix(provider): remove MiniMax /v1→/anthropic auto-correction to allow user override The minimax-specific auto-correction in runtime_provider.py was preventing users from overriding to the OpenAI-compatible endpoint via MINIMAX_BASE_URL. Users in certain regions get nginx 404 on api.minimax.io/anthropic and need to switch to api.minimax.chat/v1. The generic URL-suffix detection already handles /anthropic → anthropic_messages, so the minimax-specific code was redundant for the default path and harmful for the override path. Now: default /anthropic URL works via generic detection, user override to /v1 gets chat_completions mode naturally. Closes NousResearch#3546 (different approach — respects user overrides instead of changing the default endpoint). * fix(display): show reasoning during streaming even when tool calls suppress content When a model generates content (containing <REASONING_SCRATCHPAD> tags) alongside tool calls in the same API response, content deltas were suppressed from streaming once any tool call chunk arrived. This prevented the CLI's tag extraction from running, so reasoning was never shown during streaming. The post-response fallback then displayed reasoning AFTER the already-visible streamed response, creating a confusing reversed order. Fix: route suppressed content to stream_delta_callback even when tool calls are present. The CLI's _stream_delta handles tag extraction — reasoning tags are routed to the reasoning display box, while non-reasoning text is handled by the existing stream display logic. This ensures reasoning appears before tool execution and the final response, matching the expected visual order.

…or change (NousResearch#3574) PR NousResearch#3566 intentionally routes suppressed content to stream_delta_callback when tool calls are present, so reasoning tag extraction can fire during streaming. The test was still asserting the old behavior where content after tool calls was fully suppressed from the callback. Updated the assertion to match: content IS delivered to the callback (for tag extraction), with display-level suppression handled by the CLI's _stream_delta.