fix(anthropic_adapter): preserve reasoning_content for Kimi /coding tool-call messages#13897
Closed
HiddenPuppy wants to merge 2 commits into
Closed
fix(anthropic_adapter): preserve reasoning_content for Kimi /coding tool-call messages#13897HiddenPuppy wants to merge 2 commits into
HiddenPuppy wants to merge 2 commits into
Conversation
added 2 commits
April 22, 2026 15:18
…call messages for Kimi /coding Fixes NousResearch#13848 Kimi's /coding endpoint speaks the Anthropic Messages protocol but has its own thinking semantics: when thinking is enabled, Kimi validates message history and requires every prior assistant tool-call message to carry OpenAI-style reasoning_content. The Anthropic path never populated that field, and convert_messages_to_anthropic strips all Anthropic thinking blocks on third-party endpoints — so the request failed with HTTP 400: "thinking is enabled but reasoning_content is missing in assistant tool call message at index N" Now, when an assistant message contains tool_calls and a reasoning_content string, we append a {"type": "thinking", ...} block to the Anthropic content so Kimi can validate the history. This only affects assistant messages with tool_calls + reasoning_content; plain text assistant messages are unchanged.
Map tsuijinglei@gmail.com → hiddenpuppy.
Collaborator
kshitijk4poor
added a commit
that referenced
this pull request
Apr 22, 2026
… block ordering Follow-up to the cherry-picked PR #13897 fix. Three issues found: 1. CRITICAL: The thinking block synthesised from reasoning_content was immediately stripped by the third-party signature management code (Kimi is classified as _is_third_party_anthropic_endpoint). Added a Kimi-specific carve-out that preserves unsigned thinking blocks while still stripping Anthropic-signed blocks Kimi can't validate. 2. Empty-string reasoning_content was silently dropped because the truthiness check ('if reasoning_content and ...') evaluates to False for ''. Changed to 'isinstance(reasoning_content, str)' so the tier-3 fallback from _copy_reasoning_content_for_api (which injects '' for Kimi tool-call messages with no reasoning) actually produces a thinking block. 3. The thinking block was appended AFTER tool_use blocks. Anthropic protocol requires thinking -> text -> tool_use ordering. Changed to blocks.insert(0, ...) to prepend.
kshitijk4poor
added a commit
that referenced
this pull request
Apr 22, 2026
… block ordering Follow-up to the cherry-picked PR #13897 fix. Three issues found: 1. CRITICAL: The thinking block synthesised from reasoning_content was immediately stripped by the third-party signature management code (Kimi is classified as _is_third_party_anthropic_endpoint). Added a Kimi-specific carve-out that preserves unsigned thinking blocks while still stripping Anthropic-signed blocks Kimi can't validate. 2. Empty-string reasoning_content was silently dropped because the truthiness check ('if reasoning_content and ...') evaluates to False for ''. Changed to 'isinstance(reasoning_content, str)' so the tier-3 fallback from _copy_reasoning_content_for_api (which injects '' for Kimi tool-call messages with no reasoning) actually produces a thinking block. 3. The thinking block was appended AFTER tool_use blocks. Anthropic protocol requires thinking -> text -> tool_use ordering. Changed to blocks.insert(0, ...) to prepend.
Collaborator
|
Merged via PR #14018. Your commit was cherry-picked onto current main with your authorship preserved in git log. We added follow-up fixes on top: a Kimi-specific carve-out in the signature stripping code (your thinking block was being stripped as a third-party endpoint), empty-string reasoning_content handling, correct block ordering, and a guard to prevent duplicate thinking content on native Anthropic. Thanks for the fix! |
1 task
Contributor
Author
Thanks a lot for the merge and the detailed follow-up fixes — really appreciate the improvements you added on top. I’ve been actively contributing to the project and would love to get more involved going forward. If possible, I’d be interested in becoming a collaborator to help review PRs and contribute more efficiently. Happy to continue contributing either way — thanks again for the support! |
ulasbilgen
pushed a commit
to ulasbilgen/hermes-adhd-agent
that referenced
this pull request
May 1, 2026
… block ordering Follow-up to the cherry-picked PR NousResearch#13897 fix. Three issues found: 1. CRITICAL: The thinking block synthesised from reasoning_content was immediately stripped by the third-party signature management code (Kimi is classified as _is_third_party_anthropic_endpoint). Added a Kimi-specific carve-out that preserves unsigned thinking blocks while still stripping Anthropic-signed blocks Kimi can't validate. 2. Empty-string reasoning_content was silently dropped because the truthiness check ('if reasoning_content and ...') evaluates to False for ''. Changed to 'isinstance(reasoning_content, str)' so the tier-3 fallback from _copy_reasoning_content_for_api (which injects '' for Kimi tool-call messages with no reasoning) actually produces a thinking block. 3. The thinking block was appended AFTER tool_use blocks. Anthropic protocol requires thinking -> text -> tool_use ordering. Changed to blocks.insert(0, ...) to prepend.
aj-nt
pushed a commit
to aj-nt/hermes-agent
that referenced
this pull request
May 1, 2026
… block ordering Follow-up to the cherry-picked PR NousResearch#13897 fix. Three issues found: 1. CRITICAL: The thinking block synthesised from reasoning_content was immediately stripped by the third-party signature management code (Kimi is classified as _is_third_party_anthropic_endpoint). Added a Kimi-specific carve-out that preserves unsigned thinking blocks while still stripping Anthropic-signed blocks Kimi can't validate. 2. Empty-string reasoning_content was silently dropped because the truthiness check ('if reasoning_content and ...') evaluates to False for ''. Changed to 'isinstance(reasoning_content, str)' so the tier-3 fallback from _copy_reasoning_content_for_api (which injects '' for Kimi tool-call messages with no reasoning) actually produces a thinking block. 3. The thinking block was appended AFTER tool_use blocks. Anthropic protocol requires thinking -> text -> tool_use ordering. Changed to blocks.insert(0, ...) to prepend.
02356abc
pushed a commit
to 02356abc/hermes-agent
that referenced
this pull request
May 14, 2026
… block ordering Follow-up to the cherry-picked PR NousResearch#13897 fix. Three issues found: 1. CRITICAL: The thinking block synthesised from reasoning_content was immediately stripped by the third-party signature management code (Kimi is classified as _is_third_party_anthropic_endpoint). Added a Kimi-specific carve-out that preserves unsigned thinking blocks while still stripping Anthropic-signed blocks Kimi can't validate. 2. Empty-string reasoning_content was silently dropped because the truthiness check ('if reasoning_content and ...') evaluates to False for ''. Changed to 'isinstance(reasoning_content, str)' so the tier-3 fallback from _copy_reasoning_content_for_api (which injects '' for Kimi tool-call messages with no reasoning) actually produces a thinking block. 3. The thinking block was appended AFTER tool_use blocks. Anthropic protocol requires thinking -> text -> tool_use ordering. Changed to blocks.insert(0, ...) to prepend.
gweeteve
pushed a commit
to gweeteve/hermes-agent
that referenced
this pull request
Jun 2, 2026
… block ordering Follow-up to the cherry-picked PR NousResearch#13897 fix. Three issues found: 1. CRITICAL: The thinking block synthesised from reasoning_content was immediately stripped by the third-party signature management code (Kimi is classified as _is_third_party_anthropic_endpoint). Added a Kimi-specific carve-out that preserves unsigned thinking blocks while still stripping Anthropic-signed blocks Kimi can't validate. 2. Empty-string reasoning_content was silently dropped because the truthiness check ('if reasoning_content and ...') evaluates to False for ''. Changed to 'isinstance(reasoning_content, str)' so the tier-3 fallback from _copy_reasoning_content_for_api (which injects '' for Kimi tool-call messages with no reasoning) actually produces a thinking block. 3. The thinking block was appended AFTER tool_use blocks. Anthropic protocol requires thinking -> text -> tool_use ordering. Changed to blocks.insert(0, ...) to prepend.
Egavasyug
pushed a commit
to Egavasyug/hermes-agent
that referenced
this pull request
Jun 10, 2026
… block ordering Follow-up to the cherry-picked PR NousResearch#13897 fix. Three issues found: 1. CRITICAL: The thinking block synthesised from reasoning_content was immediately stripped by the third-party signature management code (Kimi is classified as _is_third_party_anthropic_endpoint). Added a Kimi-specific carve-out that preserves unsigned thinking blocks while still stripping Anthropic-signed blocks Kimi can't validate. 2. Empty-string reasoning_content was silently dropped because the truthiness check ('if reasoning_content and ...') evaluates to False for ''. Changed to 'isinstance(reasoning_content, str)' so the tier-3 fallback from _copy_reasoning_content_for_api (which injects '' for Kimi tool-call messages with no reasoning) actually produces a thinking block. 3. The thinking block was appended AFTER tool_use blocks. Anthropic protocol requires thinking -> text -> tool_use ordering. Changed to blocks.insert(0, ...) to prepend.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Fixes #13848
Problem
When using
kimi-for-coding(Kimi's/codingendpoint via Anthropic Messages protocol), any session that triggers tool calls becomes permanently broken after several rounds.The Moonshot API returns HTTP 400:
This happens because:
/codingendpoint speaks Anthropic Messages protocol but has its own thinking semanticsthinking.enabledis sent, Kimi validates message history and requires every prior assistant tool-call message to carry OpenAI-stylereasoning_contentconvert_messages_to_anthropicstrips all Anthropic thinking blocks on third-party endpointsFix
In
convert_messages_to_anthropic, when processing an assistant message that contains bothtool_callsandreasoning_content, preserve thereasoning_contentas a{"type": "thinking", "thinking": ...}block appended to the Anthropic content blocks.This allows Kimi to validate the message history while keeping the change minimal and only affecting messages that have both tool_calls and reasoning_content.
Changes
agent/anthropic_adapter.py:convert_messages_to_anthropic()— added reasoning_content preservation for assistant tool-call messagesVerification
pnpm check/pytestpassesChecklist