Skip to content

fix(agent): add length-continuation support for anthropic_messages mode#7242

Closed
ygd58 wants to merge 1 commit into
NousResearch:mainfrom
ygd58:fix/anthropic-length-continuation
Closed

fix(agent): add length-continuation support for anthropic_messages mode#7242
ygd58 wants to merge 1 commit into
NousResearch:mainfrom
ygd58:fix/anthropic-length-continuation

Conversation

@ygd58

@ygd58 ygd58 commented Apr 10, 2026

Copy link
Copy Markdown
Contributor

Problem

Closes #7237

When a response was truncated (finish_reason=length) in anthropic_messages mode, the agent immediately fell through to rollback or hard error. The 3-attempt auto-continuation logic existed only for chat_completions mode.

Root Cause

In run_agent.py, the continuation block was gated on if self.api_mode == "chat_completions". Anthropic native API users had no recovery path and always saw Error: Response truncated due to output length limit.

Fix

Adds a parallel continuation block for anthropic_messages mode:

  • Collects partial text from Anthropic content blocks
  • Appends interim assistant message preserving raw content blocks
  • Sends continuation prompt up to 3 times
  • Stitches chunks into a single final response
  • Returns graceful partial response after 3 failed attempts
  • Skips tool call truncations (handled by existing retry logic)

Before / After

Before: Anthropic truncation → immediate rollback + error

After: Anthropic truncation → up to 3 continuation attempts → full response delivered

@ygd58
fix(agent): add length-continuation support for anthropic_messages mode
cb2763a 
Previously, when a response was truncated (finish_reason='length') in
anthropic_messages mode, the agent immediately fell through to rollback
or error. The 3-attempt continuation logic only covered chat_completions.

This fix mirrors the existing chat_completions continuation behaviour
for anthropic_messages (Anthropic native API):
- Collects partial text from Anthropic content blocks
- Appends interim assistant message with raw content blocks
- Sends '[System: Continue...]' user message up to 3 times
- Joins all partial chunks into a single final_response
- Returns a graceful partial response after 3 failed attempts

Fixes NousResearch#7237
@ygd58 ygd58 mentioned this pull request Apr 10, 2026
Closed
1 task
teknium1 added a commit that referenced this pull request Apr 20, 2026
@teknium1 @nocoo
@keyuyuan @n-WN @sgaofen @vince-core @zhongyueming1121 @hansnow @TroyMitchell911
fix(anthropic): complete third-party Anthropic-compatible provider su…
8a0f8b1 
…pport

Third-party gateways that speak the native Anthropic protocol (MiniMax,
Zhipu GLM, Alibaba DashScope, Kimi, LiteLLM proxies) now work end-to-end
with the same feature set as direct api.anthropic.com callers.  Synthesizes
eight stale community PRs into one consolidated change.

Five fixes:

- URL detection: consolidate three inline `endswith("/anthropic")`
  checks in runtime_provider.py into the shared _detect_api_mode_for_url
  helper.  Third-party /anthropic endpoints now auto-resolve to
  api_mode=anthropic_messages via one code path instead of three.

- OAuth leak-guard: all five sites that assign `_is_anthropic_oauth`
  (__init__, switch_model, _try_refresh_anthropic_client_credentials,
  _swap_credential, _try_activate_fallback) now gate on
  `provider == "anthropic"` so a stale ANTHROPIC_TOKEN never trips
  Claude-Code identity injection on third-party endpoints.  Previously
  only 2 of 5 sites were guarded.

- Prompt caching: new method `_anthropic_prompt_cache_policy()` returns
  `(should_cache, use_native_layout)` per endpoint.  Replaces three
  inline conditions and the `native_anthropic=(api_mode=='anthropic_messages')`
  call-site flag.  Native Anthropic and third-party Anthropic gateways
  both get the native cache_control layout; OpenRouter gets envelope
  layout.  Layout is persisted in `_primary_runtime` so fallback
  restoration preserves the per-endpoint choice.

- Auxiliary client: `_try_custom_endpoint` honors
  `api_mode=anthropic_messages` and builds `AnthropicAuxiliaryClient`
  instead of silently downgrading to an OpenAI-wire client.  Degrades
  gracefully to OpenAI-wire when the anthropic SDK isn't installed.

- Config hygiene: `_update_config_for_provider` (hermes_cli/auth.py)
  clears stale `api_key`/`api_mode` when switching to a built-in
  provider, so a previous MiniMax custom endpoint's credentials can't
  leak into a later OpenRouter session.

- Truncation continuation: length-continuation and tool-call-truncation
  retry now cover `anthropic_messages` in addition to `chat_completions`
  and `bedrock_converse`.  Reuses the existing `_build_assistant_message`
  path via `normalize_anthropic_response()` so the interim message
  shape is byte-identical to the non-truncated path.

Tests: 6 new files, 42 test cases.  Targeted run + tests/run_agent,
tests/agent, tests/hermes_cli all pass (4554 passed).

Synthesized from (credits preserved via Co-authored-by trailers):
  #7410  @nocoo           — URL detection helper
  #7393  @keyuyuan        — OAuth 5-site guard
  #7367  @n-WN            — OAuth guard (narrower cousin, kept comment)
  #8636  @sgaofen         — caching helper + native-vs-proxy layout split
  #10954 @Only-Code-A     — caching on anthropic_messages+Claude
  #7648  @zhongyueming1121 — aux client anthropic_messages branch
  #6096  @hansnow         — /model switch clears stale api_mode
  #9691  @TroyMitchell911 — anthropic_messages truncation continuation

Closes: #7366, #8294 (third-party Anthropic identity + caching).
Supersedes: #7410, #7367, #7393, #8636, #10954, #7648, #6096, #9691.
Rejects:    #9621 (OpenAI-wire caching with incomplete blocklist — risky),
            #7242 (superseded by #9691, stale branch),
            #8321 (targets smart_model_routing which was removed in #12732).

Co-authored-by: nocoo <nocoo@users.noreply.github.com>
Co-authored-by: Keyu Yuan <leoyuan0099@gmail.com>
Co-authored-by: Zoee <30841158+n-WN@users.noreply.github.com>
Co-authored-by: sgaofen <135070653+sgaofen@users.noreply.github.com>
Co-authored-by: Only-Code-A <bxzt2006@163.com>
Co-authored-by: zhongyueming <mygamez@163.com>
Co-authored-by: Xiaohan Li <hansnow@users.noreply.github.com>
Co-authored-by: Troy Mitchell <i@troy-y.org>
@teknium1 teknium1 mentioned this pull request Apr 20, 2026
Merged
teknium1 added a commit that referenced this pull request Apr 20, 2026
@teknium1 @nocoo
@keyuyuan @n-WN @sgaofen @vince-core @zhongyueming1121 @hansnow @TroyMitchell911
fix(anthropic): complete third-party Anthropic-compatible provider su…
65a31ee 
…pport (#12846)

Third-party gateways that speak the native Anthropic protocol (MiniMax,
Zhipu GLM, Alibaba DashScope, Kimi, LiteLLM proxies) now work end-to-end
with the same feature set as direct api.anthropic.com callers.  Synthesizes
eight stale community PRs into one consolidated change.

Five fixes:

- URL detection: consolidate three inline `endswith("/anthropic")`
  checks in runtime_provider.py into the shared _detect_api_mode_for_url
  helper.  Third-party /anthropic endpoints now auto-resolve to
  api_mode=anthropic_messages via one code path instead of three.

- OAuth leak-guard: all five sites that assign `_is_anthropic_oauth`
  (__init__, switch_model, _try_refresh_anthropic_client_credentials,
  _swap_credential, _try_activate_fallback) now gate on
  `provider == "anthropic"` so a stale ANTHROPIC_TOKEN never trips
  Claude-Code identity injection on third-party endpoints.  Previously
  only 2 of 5 sites were guarded.

- Prompt caching: new method `_anthropic_prompt_cache_policy()` returns
  `(should_cache, use_native_layout)` per endpoint.  Replaces three
  inline conditions and the `native_anthropic=(api_mode=='anthropic_messages')`
  call-site flag.  Native Anthropic and third-party Anthropic gateways
  both get the native cache_control layout; OpenRouter gets envelope
  layout.  Layout is persisted in `_primary_runtime` so fallback
  restoration preserves the per-endpoint choice.

- Auxiliary client: `_try_custom_endpoint` honors
  `api_mode=anthropic_messages` and builds `AnthropicAuxiliaryClient`
  instead of silently downgrading to an OpenAI-wire client.  Degrades
  gracefully to OpenAI-wire when the anthropic SDK isn't installed.

- Config hygiene: `_update_config_for_provider` (hermes_cli/auth.py)
  clears stale `api_key`/`api_mode` when switching to a built-in
  provider, so a previous MiniMax custom endpoint's credentials can't
  leak into a later OpenRouter session.

- Truncation continuation: length-continuation and tool-call-truncation
  retry now cover `anthropic_messages` in addition to `chat_completions`
  and `bedrock_converse`.  Reuses the existing `_build_assistant_message`
  path via `normalize_anthropic_response()` so the interim message
  shape is byte-identical to the non-truncated path.

Tests: 6 new files, 42 test cases.  Targeted run + tests/run_agent,
tests/agent, tests/hermes_cli all pass (4554 passed).

Synthesized from (credits preserved via Co-authored-by trailers):
  #7410  @nocoo           — URL detection helper
  #7393  @keyuyuan        — OAuth 5-site guard
  #7367  @n-WN            — OAuth guard (narrower cousin, kept comment)
  #8636  @sgaofen         — caching helper + native-vs-proxy layout split
  #10954 @Only-Code-A     — caching on anthropic_messages+Claude
  #7648  @zhongyueming1121 — aux client anthropic_messages branch
  #6096  @hansnow         — /model switch clears stale api_mode
  #9691  @TroyMitchell911 — anthropic_messages truncation continuation

Closes: #7366, #8294 (third-party Anthropic identity + caching).
Supersedes: #7410, #7367, #7393, #8636, #10954, #7648, #6096, #9691.
Rejects:    #9621 (OpenAI-wire caching with incomplete blocklist — risky),
            #7242 (superseded by #9691, stale branch),
            #8321 (targets smart_model_routing which was removed in #12732).

Co-authored-by: nocoo <nocoo@users.noreply.github.com>
Co-authored-by: Keyu Yuan <leoyuan0099@gmail.com>
Co-authored-by: Zoee <30841158+n-WN@users.noreply.github.com>
Co-authored-by: sgaofen <135070653+sgaofen@users.noreply.github.com>
Co-authored-by: Only-Code-A <bxzt2006@163.com>
Co-authored-by: zhongyueming <mygamez@163.com>
Co-authored-by: Xiaohan Li <hansnow@users.noreply.github.com>
Co-authored-by: Troy Mitchell <i@troy-y.org>
@teknium1

teknium1 commented Apr 20, 2026

Copy link
Copy Markdown
Contributor

Superseded by PR #9691 / PR #12846. PR #9691 took a cleaner approach (reusing the normalize_anthropic_response() adapter) and also covered the tool-call truncation retry path that this PR explicitly skipped. Thanks for raising this — the final fix ships in PR #12846.

Fix shipped in: #12846

@teknium1 teknium1 closed this Apr 20, 2026
ulasbilgen pushed a commit to ulasbilgen/hermes-adhd-agent that referenced this pull request May 1, 2026
@teknium1 @nocoo
@keyuyuan @n-WN @sgaofen @vince-core @zhongyueming1121 @hansnow @TroyMitchell911
fix(anthropic): complete third-party Anthropic-compatible provider su…
bd0dab3 
…pport (NousResearch#12846)

Third-party gateways that speak the native Anthropic protocol (MiniMax,
Zhipu GLM, Alibaba DashScope, Kimi, LiteLLM proxies) now work end-to-end
with the same feature set as direct api.anthropic.com callers.  Synthesizes
eight stale community PRs into one consolidated change.

Five fixes:

- URL detection: consolidate three inline `endswith("/anthropic")`
  checks in runtime_provider.py into the shared _detect_api_mode_for_url
  helper.  Third-party /anthropic endpoints now auto-resolve to
  api_mode=anthropic_messages via one code path instead of three.

- OAuth leak-guard: all five sites that assign `_is_anthropic_oauth`
  (__init__, switch_model, _try_refresh_anthropic_client_credentials,
  _swap_credential, _try_activate_fallback) now gate on
  `provider == "anthropic"` so a stale ANTHROPIC_TOKEN never trips
  Claude-Code identity injection on third-party endpoints.  Previously
  only 2 of 5 sites were guarded.

- Prompt caching: new method `_anthropic_prompt_cache_policy()` returns
  `(should_cache, use_native_layout)` per endpoint.  Replaces three
  inline conditions and the `native_anthropic=(api_mode=='anthropic_messages')`
  call-site flag.  Native Anthropic and third-party Anthropic gateways
  both get the native cache_control layout; OpenRouter gets envelope
  layout.  Layout is persisted in `_primary_runtime` so fallback
  restoration preserves the per-endpoint choice.

- Auxiliary client: `_try_custom_endpoint` honors
  `api_mode=anthropic_messages` and builds `AnthropicAuxiliaryClient`
  instead of silently downgrading to an OpenAI-wire client.  Degrades
  gracefully to OpenAI-wire when the anthropic SDK isn't installed.

- Config hygiene: `_update_config_for_provider` (hermes_cli/auth.py)
  clears stale `api_key`/`api_mode` when switching to a built-in
  provider, so a previous MiniMax custom endpoint's credentials can't
  leak into a later OpenRouter session.

- Truncation continuation: length-continuation and tool-call-truncation
  retry now cover `anthropic_messages` in addition to `chat_completions`
  and `bedrock_converse`.  Reuses the existing `_build_assistant_message`
  path via `normalize_anthropic_response()` so the interim message
  shape is byte-identical to the non-truncated path.

Tests: 6 new files, 42 test cases.  Targeted run + tests/run_agent,
tests/agent, tests/hermes_cli all pass (4554 passed).

Synthesized from (credits preserved via Co-authored-by trailers):
  NousResearch#7410  @nocoo           — URL detection helper
  NousResearch#7393  @keyuyuan        — OAuth 5-site guard
  NousResearch#7367  @n-WN            — OAuth guard (narrower cousin, kept comment)
  NousResearch#8636  @sgaofen         — caching helper + native-vs-proxy layout split
  NousResearch#10954 @Only-Code-A     — caching on anthropic_messages+Claude
  NousResearch#7648  @zhongyueming1121 — aux client anthropic_messages branch
  NousResearch#6096  @hansnow         — /model switch clears stale api_mode
  NousResearch#9691  @TroyMitchell911 — anthropic_messages truncation continuation

Closes: NousResearch#7366, NousResearch#8294 (third-party Anthropic identity + caching).
Supersedes: NousResearch#7410, NousResearch#7367, NousResearch#7393, NousResearch#8636, NousResearch#10954, NousResearch#7648, NousResearch#6096, NousResearch#9691.
Rejects:    NousResearch#9621 (OpenAI-wire caching with incomplete blocklist — risky),
            NousResearch#7242 (superseded by NousResearch#9691, stale branch),
            NousResearch#8321 (targets smart_model_routing which was removed in NousResearch#12732).

Co-authored-by: nocoo <nocoo@users.noreply.github.com>
Co-authored-by: Keyu Yuan <leoyuan0099@gmail.com>
Co-authored-by: Zoee <30841158+n-WN@users.noreply.github.com>
Co-authored-by: sgaofen <135070653+sgaofen@users.noreply.github.com>
Co-authored-by: Only-Code-A <bxzt2006@163.com>
Co-authored-by: zhongyueming <mygamez@163.com>
Co-authored-by: Xiaohan Li <hansnow@users.noreply.github.com>
Co-authored-by: Troy Mitchell <i@troy-y.org>
aj-nt pushed a commit to aj-nt/hermes-agent that referenced this pull request May 1, 2026
@teknium1 @nocoo
@keyuyuan @n-WN @sgaofen @vince-core @zhongyueming1121 @hansnow @TroyMitchell911
fix(anthropic): complete third-party Anthropic-compatible provider su…
213769a 
…pport (NousResearch#12846)

Third-party gateways that speak the native Anthropic protocol (MiniMax,
Zhipu GLM, Alibaba DashScope, Kimi, LiteLLM proxies) now work end-to-end
with the same feature set as direct api.anthropic.com callers.  Synthesizes
eight stale community PRs into one consolidated change.

Five fixes:

- URL detection: consolidate three inline `endswith("/anthropic")`
  checks in runtime_provider.py into the shared _detect_api_mode_for_url
  helper.  Third-party /anthropic endpoints now auto-resolve to
  api_mode=anthropic_messages via one code path instead of three.

- OAuth leak-guard: all five sites that assign `_is_anthropic_oauth`
  (__init__, switch_model, _try_refresh_anthropic_client_credentials,
  _swap_credential, _try_activate_fallback) now gate on
  `provider == "anthropic"` so a stale ANTHROPIC_TOKEN never trips
  Claude-Code identity injection on third-party endpoints.  Previously
  only 2 of 5 sites were guarded.

- Prompt caching: new method `_anthropic_prompt_cache_policy()` returns
  `(should_cache, use_native_layout)` per endpoint.  Replaces three
  inline conditions and the `native_anthropic=(api_mode=='anthropic_messages')`
  call-site flag.  Native Anthropic and third-party Anthropic gateways
  both get the native cache_control layout; OpenRouter gets envelope
  layout.  Layout is persisted in `_primary_runtime` so fallback
  restoration preserves the per-endpoint choice.

- Auxiliary client: `_try_custom_endpoint` honors
  `api_mode=anthropic_messages` and builds `AnthropicAuxiliaryClient`
  instead of silently downgrading to an OpenAI-wire client.  Degrades
  gracefully to OpenAI-wire when the anthropic SDK isn't installed.

- Config hygiene: `_update_config_for_provider` (hermes_cli/auth.py)
  clears stale `api_key`/`api_mode` when switching to a built-in
  provider, so a previous MiniMax custom endpoint's credentials can't
  leak into a later OpenRouter session.

- Truncation continuation: length-continuation and tool-call-truncation
  retry now cover `anthropic_messages` in addition to `chat_completions`
  and `bedrock_converse`.  Reuses the existing `_build_assistant_message`
  path via `normalize_anthropic_response()` so the interim message
  shape is byte-identical to the non-truncated path.

Tests: 6 new files, 42 test cases.  Targeted run + tests/run_agent,
tests/agent, tests/hermes_cli all pass (4554 passed).

Synthesized from (credits preserved via Co-authored-by trailers):
  NousResearch#7410  @nocoo           — URL detection helper
  NousResearch#7393  @keyuyuan        — OAuth 5-site guard
  NousResearch#7367  @n-WN            — OAuth guard (narrower cousin, kept comment)
  NousResearch#8636  @sgaofen         — caching helper + native-vs-proxy layout split
  NousResearch#10954 @Only-Code-A     — caching on anthropic_messages+Claude
  NousResearch#7648  @zhongyueming1121 — aux client anthropic_messages branch
  NousResearch#6096  @hansnow         — /model switch clears stale api_mode
  NousResearch#9691  @TroyMitchell911 — anthropic_messages truncation continuation

Closes: NousResearch#7366, NousResearch#8294 (third-party Anthropic identity + caching).
Supersedes: NousResearch#7410, NousResearch#7367, NousResearch#7393, NousResearch#8636, NousResearch#10954, NousResearch#7648, NousResearch#6096, NousResearch#9691.
Rejects:    NousResearch#9621 (OpenAI-wire caching with incomplete blocklist — risky),
            NousResearch#7242 (superseded by NousResearch#9691, stale branch),
            NousResearch#8321 (targets smart_model_routing which was removed in NousResearch#12732).

Co-authored-by: nocoo <nocoo@users.noreply.github.com>
Co-authored-by: Keyu Yuan <leoyuan0099@gmail.com>
Co-authored-by: Zoee <30841158+n-WN@users.noreply.github.com>
Co-authored-by: sgaofen <135070653+sgaofen@users.noreply.github.com>
Co-authored-by: Only-Code-A <bxzt2006@163.com>
Co-authored-by: zhongyueming <mygamez@163.com>
Co-authored-by: Xiaohan Li <hansnow@users.noreply.github.com>
Co-authored-by: Troy Mitchell <i@troy-y.org>
Luminet2023 pushed a commit to Luminet2023/hermes-agent that referenced this pull request May 1, 2026
@teknium1 @nocoo
@keyuyuan @n-WN @sgaofen @vince-core @zhongyueming1121 @hansnow @TroyMitchell911
fix(anthropic): complete third-party Anthropic-compatible provider su…
dea97af 
…pport (NousResearch#12846)

Third-party gateways that speak the native Anthropic protocol (MiniMax,
Zhipu GLM, Alibaba DashScope, Kimi, LiteLLM proxies) now work end-to-end
with the same feature set as direct api.anthropic.com callers.  Synthesizes
eight stale community PRs into one consolidated change.

Five fixes:

- URL detection: consolidate three inline `endswith("/anthropic")`
  checks in runtime_provider.py into the shared _detect_api_mode_for_url
  helper.  Third-party /anthropic endpoints now auto-resolve to
  api_mode=anthropic_messages via one code path instead of three.

- OAuth leak-guard: all five sites that assign `_is_anthropic_oauth`
  (__init__, switch_model, _try_refresh_anthropic_client_credentials,
  _swap_credential, _try_activate_fallback) now gate on
  `provider == "anthropic"` so a stale ANTHROPIC_TOKEN never trips
  Claude-Code identity injection on third-party endpoints.  Previously
  only 2 of 5 sites were guarded.

- Prompt caching: new method `_anthropic_prompt_cache_policy()` returns
  `(should_cache, use_native_layout)` per endpoint.  Replaces three
  inline conditions and the `native_anthropic=(api_mode=='anthropic_messages')`
  call-site flag.  Native Anthropic and third-party Anthropic gateways
  both get the native cache_control layout; OpenRouter gets envelope
  layout.  Layout is persisted in `_primary_runtime` so fallback
  restoration preserves the per-endpoint choice.

- Auxiliary client: `_try_custom_endpoint` honors
  `api_mode=anthropic_messages` and builds `AnthropicAuxiliaryClient`
  instead of silently downgrading to an OpenAI-wire client.  Degrades
  gracefully to OpenAI-wire when the anthropic SDK isn't installed.

- Config hygiene: `_update_config_for_provider` (hermes_cli/auth.py)
  clears stale `api_key`/`api_mode` when switching to a built-in
  provider, so a previous MiniMax custom endpoint's credentials can't
  leak into a later OpenRouter session.

- Truncation continuation: length-continuation and tool-call-truncation
  retry now cover `anthropic_messages` in addition to `chat_completions`
  and `bedrock_converse`.  Reuses the existing `_build_assistant_message`
  path via `normalize_anthropic_response()` so the interim message
  shape is byte-identical to the non-truncated path.

Tests: 6 new files, 42 test cases.  Targeted run + tests/run_agent,
tests/agent, tests/hermes_cli all pass (4554 passed).

Synthesized from (credits preserved via Co-authored-by trailers):
  NousResearch#7410  @nocoo           — URL detection helper
  NousResearch#7393  @keyuyuan        — OAuth 5-site guard
  NousResearch#7367  @n-WN            — OAuth guard (narrower cousin, kept comment)
  NousResearch#8636  @sgaofen         — caching helper + native-vs-proxy layout split
  NousResearch#10954 @Only-Code-A     — caching on anthropic_messages+Claude
  NousResearch#7648  @zhongyueming1121 — aux client anthropic_messages branch
  NousResearch#6096  @hansnow         — /model switch clears stale api_mode
  NousResearch#9691  @TroyMitchell911 — anthropic_messages truncation continuation

Closes: NousResearch#7366, NousResearch#8294 (third-party Anthropic identity + caching).
Supersedes: NousResearch#7410, NousResearch#7367, NousResearch#7393, NousResearch#8636, NousResearch#10954, NousResearch#7648, NousResearch#6096, NousResearch#9691.
Rejects:    NousResearch#9621 (OpenAI-wire caching with incomplete blocklist — risky),
            NousResearch#7242 (superseded by NousResearch#9691, stale branch),
            NousResearch#8321 (targets smart_model_routing which was removed in NousResearch#12732).

Co-authored-by: nocoo <nocoo@users.noreply.github.com>
Co-authored-by: Keyu Yuan <leoyuan0099@gmail.com>
Co-authored-by: Zoee <30841158+n-WN@users.noreply.github.com>
Co-authored-by: sgaofen <135070653+sgaofen@users.noreply.github.com>
Co-authored-by: Only-Code-A <bxzt2006@163.com>
Co-authored-by: zhongyueming <mygamez@163.com>
Co-authored-by: Xiaohan Li <hansnow@users.noreply.github.com>
Co-authored-by: Troy Mitchell <i@troy-y.org>
02356abc pushed a commit to 02356abc/hermes-agent that referenced this pull request May 14, 2026
@teknium1 @nocoo
@keyuyuan @n-WN @sgaofen @vince-core @zhongyueming1121 @hansnow @TroyMitchell911
fix(anthropic): complete third-party Anthropic-compatible provider su…
bf43e46 
…pport (NousResearch#12846)

Third-party gateways that speak the native Anthropic protocol (MiniMax,
Zhipu GLM, Alibaba DashScope, Kimi, LiteLLM proxies) now work end-to-end
with the same feature set as direct api.anthropic.com callers.  Synthesizes
eight stale community PRs into one consolidated change.

Five fixes:

- URL detection: consolidate three inline `endswith("/anthropic")`
  checks in runtime_provider.py into the shared _detect_api_mode_for_url
  helper.  Third-party /anthropic endpoints now auto-resolve to
  api_mode=anthropic_messages via one code path instead of three.

- OAuth leak-guard: all five sites that assign `_is_anthropic_oauth`
  (__init__, switch_model, _try_refresh_anthropic_client_credentials,
  _swap_credential, _try_activate_fallback) now gate on
  `provider == "anthropic"` so a stale ANTHROPIC_TOKEN never trips
  Claude-Code identity injection on third-party endpoints.  Previously
  only 2 of 5 sites were guarded.

- Prompt caching: new method `_anthropic_prompt_cache_policy()` returns
  `(should_cache, use_native_layout)` per endpoint.  Replaces three
  inline conditions and the `native_anthropic=(api_mode=='anthropic_messages')`
  call-site flag.  Native Anthropic and third-party Anthropic gateways
  both get the native cache_control layout; OpenRouter gets envelope
  layout.  Layout is persisted in `_primary_runtime` so fallback
  restoration preserves the per-endpoint choice.

- Auxiliary client: `_try_custom_endpoint` honors
  `api_mode=anthropic_messages` and builds `AnthropicAuxiliaryClient`
  instead of silently downgrading to an OpenAI-wire client.  Degrades
  gracefully to OpenAI-wire when the anthropic SDK isn't installed.

- Config hygiene: `_update_config_for_provider` (hermes_cli/auth.py)
  clears stale `api_key`/`api_mode` when switching to a built-in
  provider, so a previous MiniMax custom endpoint's credentials can't
  leak into a later OpenRouter session.

- Truncation continuation: length-continuation and tool-call-truncation
  retry now cover `anthropic_messages` in addition to `chat_completions`
  and `bedrock_converse`.  Reuses the existing `_build_assistant_message`
  path via `normalize_anthropic_response()` so the interim message
  shape is byte-identical to the non-truncated path.

Tests: 6 new files, 42 test cases.  Targeted run + tests/run_agent,
tests/agent, tests/hermes_cli all pass (4554 passed).

Synthesized from (credits preserved via Co-authored-by trailers):
  NousResearch#7410  @nocoo           — URL detection helper
  NousResearch#7393  @keyuyuan        — OAuth 5-site guard
  NousResearch#7367  @n-WN            — OAuth guard (narrower cousin, kept comment)
  NousResearch#8636  @sgaofen         — caching helper + native-vs-proxy layout split
  NousResearch#10954 @Only-Code-A     — caching on anthropic_messages+Claude
  NousResearch#7648  @zhongyueming1121 — aux client anthropic_messages branch
  NousResearch#6096  @hansnow         — /model switch clears stale api_mode
  NousResearch#9691  @TroyMitchell911 — anthropic_messages truncation continuation

Closes: NousResearch#7366, NousResearch#8294 (third-party Anthropic identity + caching).
Supersedes: NousResearch#7410, NousResearch#7367, NousResearch#7393, NousResearch#8636, NousResearch#10954, NousResearch#7648, NousResearch#6096, NousResearch#9691.
Rejects:    NousResearch#9621 (OpenAI-wire caching with incomplete blocklist — risky),
            NousResearch#7242 (superseded by NousResearch#9691, stale branch),
            NousResearch#8321 (targets smart_model_routing which was removed in NousResearch#12732).

Co-authored-by: nocoo <nocoo@users.noreply.github.com>
Co-authored-by: Keyu Yuan <leoyuan0099@gmail.com>
Co-authored-by: Zoee <30841158+n-WN@users.noreply.github.com>
Co-authored-by: sgaofen <135070653+sgaofen@users.noreply.github.com>
Co-authored-by: Only-Code-A <bxzt2006@163.com>
Co-authored-by: zhongyueming <mygamez@163.com>
Co-authored-by: Xiaohan Li <hansnow@users.noreply.github.com>
Co-authored-by: Troy Mitchell <i@troy-y.org>
@aurelienp-alt aurelienp-alt mentioned this pull request May 15, 2026
Closed
1 task
gweeteve pushed a commit to gweeteve/hermes-agent that referenced this pull request Jun 2, 2026
@teknium1 @nocoo
@keyuyuan @n-WN @sgaofen @vince-core @zhongyueming1121 @hansnow @TroyMitchell911
fix(anthropic): complete third-party Anthropic-compatible provider su…
c89fb91 
…pport (NousResearch#12846)

Third-party gateways that speak the native Anthropic protocol (MiniMax,
Zhipu GLM, Alibaba DashScope, Kimi, LiteLLM proxies) now work end-to-end
with the same feature set as direct api.anthropic.com callers.  Synthesizes
eight stale community PRs into one consolidated change.

Five fixes:

- URL detection: consolidate three inline `endswith("/anthropic")`
  checks in runtime_provider.py into the shared _detect_api_mode_for_url
  helper.  Third-party /anthropic endpoints now auto-resolve to
  api_mode=anthropic_messages via one code path instead of three.

- OAuth leak-guard: all five sites that assign `_is_anthropic_oauth`
  (__init__, switch_model, _try_refresh_anthropic_client_credentials,
  _swap_credential, _try_activate_fallback) now gate on
  `provider == "anthropic"` so a stale ANTHROPIC_TOKEN never trips
  Claude-Code identity injection on third-party endpoints.  Previously
  only 2 of 5 sites were guarded.

- Prompt caching: new method `_anthropic_prompt_cache_policy()` returns
  `(should_cache, use_native_layout)` per endpoint.  Replaces three
  inline conditions and the `native_anthropic=(api_mode=='anthropic_messages')`
  call-site flag.  Native Anthropic and third-party Anthropic gateways
  both get the native cache_control layout; OpenRouter gets envelope
  layout.  Layout is persisted in `_primary_runtime` so fallback
  restoration preserves the per-endpoint choice.

- Auxiliary client: `_try_custom_endpoint` honors
  `api_mode=anthropic_messages` and builds `AnthropicAuxiliaryClient`
  instead of silently downgrading to an OpenAI-wire client.  Degrades
  gracefully to OpenAI-wire when the anthropic SDK isn't installed.

- Config hygiene: `_update_config_for_provider` (hermes_cli/auth.py)
  clears stale `api_key`/`api_mode` when switching to a built-in
  provider, so a previous MiniMax custom endpoint's credentials can't
  leak into a later OpenRouter session.

- Truncation continuation: length-continuation and tool-call-truncation
  retry now cover `anthropic_messages` in addition to `chat_completions`
  and `bedrock_converse`.  Reuses the existing `_build_assistant_message`
  path via `normalize_anthropic_response()` so the interim message
  shape is byte-identical to the non-truncated path.

Tests: 6 new files, 42 test cases.  Targeted run + tests/run_agent,
tests/agent, tests/hermes_cli all pass (4554 passed).

Synthesized from (credits preserved via Co-authored-by trailers):
  NousResearch#7410  @nocoo           — URL detection helper
  NousResearch#7393  @keyuyuan        — OAuth 5-site guard
  NousResearch#7367  @n-WN            — OAuth guard (narrower cousin, kept comment)
  NousResearch#8636  @sgaofen         — caching helper + native-vs-proxy layout split
  NousResearch#10954 @Only-Code-A     — caching on anthropic_messages+Claude
  NousResearch#7648  @zhongyueming1121 — aux client anthropic_messages branch
  NousResearch#6096  @hansnow         — /model switch clears stale api_mode
  NousResearch#9691  @TroyMitchell911 — anthropic_messages truncation continuation

Closes: NousResearch#7366, NousResearch#8294 (third-party Anthropic identity + caching).
Supersedes: NousResearch#7410, NousResearch#7367, NousResearch#7393, NousResearch#8636, NousResearch#10954, NousResearch#7648, NousResearch#6096, NousResearch#9691.
Rejects:    NousResearch#9621 (OpenAI-wire caching with incomplete blocklist — risky),
            NousResearch#7242 (superseded by NousResearch#9691, stale branch),
            NousResearch#8321 (targets smart_model_routing which was removed in NousResearch#12732).

Co-authored-by: nocoo <nocoo@users.noreply.github.com>
Co-authored-by: Keyu Yuan <leoyuan0099@gmail.com>
Co-authored-by: Zoee <30841158+n-WN@users.noreply.github.com>
Co-authored-by: sgaofen <135070653+sgaofen@users.noreply.github.com>
Co-authored-by: Only-Code-A <bxzt2006@163.com>
Co-authored-by: zhongyueming <mygamez@163.com>
Co-authored-by: Xiaohan Li <hansnow@users.noreply.github.com>
Co-authored-by: Troy Mitchell <i@troy-y.org>
Egavasyug pushed a commit to Egavasyug/hermes-agent that referenced this pull request Jun 10, 2026
@teknium1 @nocoo
@keyuyuan @n-WN @sgaofen @vince-core @zhongyueming1121 @hansnow @TroyMitchell911
fix(anthropic): complete third-party Anthropic-compatible provider su…
4c90ef1 
…pport (NousResearch#12846)

Third-party gateways that speak the native Anthropic protocol (MiniMax,
Zhipu GLM, Alibaba DashScope, Kimi, LiteLLM proxies) now work end-to-end
with the same feature set as direct api.anthropic.com callers.  Synthesizes
eight stale community PRs into one consolidated change.

Five fixes:

- URL detection: consolidate three inline `endswith("/anthropic")`
  checks in runtime_provider.py into the shared _detect_api_mode_for_url
  helper.  Third-party /anthropic endpoints now auto-resolve to
  api_mode=anthropic_messages via one code path instead of three.

- OAuth leak-guard: all five sites that assign `_is_anthropic_oauth`
  (__init__, switch_model, _try_refresh_anthropic_client_credentials,
  _swap_credential, _try_activate_fallback) now gate on
  `provider == "anthropic"` so a stale ANTHROPIC_TOKEN never trips
  Claude-Code identity injection on third-party endpoints.  Previously
  only 2 of 5 sites were guarded.

- Prompt caching: new method `_anthropic_prompt_cache_policy()` returns
  `(should_cache, use_native_layout)` per endpoint.  Replaces three
  inline conditions and the `native_anthropic=(api_mode=='anthropic_messages')`
  call-site flag.  Native Anthropic and third-party Anthropic gateways
  both get the native cache_control layout; OpenRouter gets envelope
  layout.  Layout is persisted in `_primary_runtime` so fallback
  restoration preserves the per-endpoint choice.

- Auxiliary client: `_try_custom_endpoint` honors
  `api_mode=anthropic_messages` and builds `AnthropicAuxiliaryClient`
  instead of silently downgrading to an OpenAI-wire client.  Degrades
  gracefully to OpenAI-wire when the anthropic SDK isn't installed.

- Config hygiene: `_update_config_for_provider` (hermes_cli/auth.py)
  clears stale `api_key`/`api_mode` when switching to a built-in
  provider, so a previous MiniMax custom endpoint's credentials can't
  leak into a later OpenRouter session.

- Truncation continuation: length-continuation and tool-call-truncation
  retry now cover `anthropic_messages` in addition to `chat_completions`
  and `bedrock_converse`.  Reuses the existing `_build_assistant_message`
  path via `normalize_anthropic_response()` so the interim message
  shape is byte-identical to the non-truncated path.

Tests: 6 new files, 42 test cases.  Targeted run + tests/run_agent,
tests/agent, tests/hermes_cli all pass (4554 passed).

Synthesized from (credits preserved via Co-authored-by trailers):
  NousResearch#7410  @nocoo           — URL detection helper
  NousResearch#7393  @keyuyuan        — OAuth 5-site guard
  NousResearch#7367  @n-WN            — OAuth guard (narrower cousin, kept comment)
  NousResearch#8636  @sgaofen         — caching helper + native-vs-proxy layout split
  NousResearch#10954 @Only-Code-A     — caching on anthropic_messages+Claude
  NousResearch#7648  @zhongyueming1121 — aux client anthropic_messages branch
  NousResearch#6096  @hansnow         — /model switch clears stale api_mode
  NousResearch#9691  @TroyMitchell911 — anthropic_messages truncation continuation

Closes: NousResearch#7366, NousResearch#8294 (third-party Anthropic identity + caching).
Supersedes: NousResearch#7410, NousResearch#7367, NousResearch#7393, NousResearch#8636, NousResearch#10954, NousResearch#7648, NousResearch#6096, NousResearch#9691.
Rejects:    NousResearch#9621 (OpenAI-wire caching with incomplete blocklist — risky),
            NousResearch#7242 (superseded by NousResearch#9691, stale branch),
            NousResearch#8321 (targets smart_model_routing which was removed in NousResearch#12732).

Co-authored-by: nocoo <nocoo@users.noreply.github.com>
Co-authored-by: Keyu Yuan <leoyuan0099@gmail.com>
Co-authored-by: Zoee <30841158+n-WN@users.noreply.github.com>
Co-authored-by: sgaofen <135070653+sgaofen@users.noreply.github.com>
Co-authored-by: Only-Code-A <bxzt2006@163.com>
Co-authored-by: zhongyueming <mygamez@163.com>
Co-authored-by: Xiaohan Li <hansnow@users.noreply.github.com>
Co-authored-by: Troy Mitchell <i@troy-y.org>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

[Bug]: Error: Response truncated due to output length limit

2 participants

@ygd58 @teknium1