Skip to content

fix(xai-oauth): strip service_tier and add safety-net sanitization for slash enums#28490

Closed
Nami4D wants to merge 1 commit into
NousResearch:mainfrom
Nami4D:fix/xai-oauth-invalid-arguments
Closed

fix(xai-oauth): strip service_tier and add safety-net sanitization for slash enums#28490
Nami4D wants to merge 1 commit into
NousResearch:mainfrom
Nami4D:fix/xai-oauth-invalid-arguments

Conversation

@Nami4D

@Nami4D Nami4D commented May 19, 2026

Copy link
Copy Markdown
Contributor

What changed

Strip service_tier from xAI Responses API requests and add a safety-net strip_slash_enum call in the preflight path.

Why

xAI's /v1/responses endpoint rejects two request shapes with HTTP 400:

  1. service_tier — "Argument not supported: service_tier" (hit when /fast mode is active)
  2. Tool schema enum values containing / — "Invalid arguments passed to the model" (HuggingFace model IDs like Qwen/Qwen3.5-0.8B from MCP servers). The existing strip_slash_enum sanitizer in build_api_kwargs is bypassed in some code paths, so this adds a safety-net call in the preflight function.

How to test

Run Hermes with xai-oauth provider (SuperGrok):

hermes model  # → pick xAI Grok OAuth (SuperGrok Subscription)
hermes        # → chat normally

Platforms tested

  • macOS (direct API testing with 23 request variants against api.x.ai/v1/responses + end-to-end Hermes agent runs)

Tests

  • tests/agent/transports/test_codex_transport.py — 39 passed
  • tests/run_agent/test_run_agent_codex_responses.py — 63 passed
  • scripts/check-windows-footguns.py — clean

@Nami4D
fix(xai-oauth): strip service_tier and add safety-net sanitization fo…
7ea58fd 
…r slash enums

xAI's /v1/responses endpoint rejects service_tier with HTTP 400
"Argument not supported: service_tier" when users activate /fast mode.

Also add a safety-net strip_slash_enum call in _preflight_codex_api_kwargs
to catch any tool schemas that might slip through the caller-level
sanitization. xAI's Responses API grammar compiler rejects enum values
containing forward slashes (e.g. HuggingFace model IDs like
"Qwen/Qwen3.5-0.8B") with the opaque "Invalid arguments passed to the
model" error.

Fixes the root cause of "Invalid arguments passed to the model" errors
reported by xAI OAuth (SuperGrok) users.
@alt-glitch alt-glitch added type/bug Something isn't working P3 Low — cosmetic, nice to have comp/agent Core agent loop, run_agent.py, prompt builder provider/xai xAI (Grok) labels May 19, 2026
@teknium1 teknium1 mentioned this pull request May 26, 2026
Merged
teknium1 added a commit that referenced this pull request May 27, 2026
@teknium1 @Nami4D
fix(xai-oauth): gate slash-enum strip on model name + add regression …
b4eea18 
…tests (#28490)

Three additions on top of @Nami4D's salvage:

1. Gate the preflight slash-enum strip on the model name pattern
   (grok-* / x-ai/grok-*).  The original PR stripped slash-containing
   enum values from every codex_responses request, but native Codex
   (OpenAI) and GitHub Models DO accept slash enums — stripping them
   there would silently degrade tool-schema constraints.  xAI is the
   only Responses-API surface that rejects the shape.

2. Resolve the merge conflict in agent/transports/codex.py by
   preserving both the timeout-forwarding block that landed on main
   between the PR's branch point and now AND the new service_tier
   strip.  Behavioural intent of both is preserved.

3. Six new tests in tests/agent/transports/test_codex_transport.py
   covering:
   - TestCodexTransportXaiServiceTierStrip (3 tests): xAI strips
     service_tier from request_overrides; non-xAI codex_responses
     and GitHub Models both KEEP service_tier (regression guards
     so the strip stays xAI-only).
   - TestPreflightSlashEnumStrip (3 tests): Grok and aggregator-
     prefixed Grok model names both trigger the safety-net strip;
     non-Grok models preserve slash enums as a regression guard
     against the strip becoming too broad.

51/51 in tests/agent/transports/test_codex_transport.py.

Co-authored-by: Nami4D <hello@nami4d.tech>
@teknium1

teknium1 commented May 27, 2026

Copy link
Copy Markdown
Contributor

Closing — merged via salvage PR #32443 (#32443).

Your two commits were cherry-picked onto current main with authorship preserved (Co-authored-by: trailer in the merge commit). The salvage added:

  • Resolution for a merge conflict from a timeout-forwarding block that landed between your branch point and main.
  • A model-name gate on the slash-enum preflight strip so it only fires for Grok models (the original PR applied to all codex_responses requests, but native Codex / GitHub Models actually accept slash enums and stripping the constraint there would silently degrade tool-schema correctness).
  • 6 regression tests covering both fixes in both directions.

Both bugs are now fixed on main. Thanks for the careful diagnosis — 23 request variants against api.x.ai is exactly the empirical work that makes these salvageable in an afternoon.

@teknium1 teknium1 closed this May 27, 2026
mathias3 pushed a commit to mathias3/hermes-agent that referenced this pull request May 28, 2026
@teknium1 @Nami4D @mathias3
fix(xai-oauth): gate slash-enum strip on model name + add regression …
8ee4b88 
…tests (NousResearch#28490)

Three additions on top of @Nami4D's salvage:

1. Gate the preflight slash-enum strip on the model name pattern
   (grok-* / x-ai/grok-*).  The original PR stripped slash-containing
   enum values from every codex_responses request, but native Codex
   (OpenAI) and GitHub Models DO accept slash enums — stripping them
   there would silently degrade tool-schema constraints.  xAI is the
   only Responses-API surface that rejects the shape.

2. Resolve the merge conflict in agent/transports/codex.py by
   preserving both the timeout-forwarding block that landed on main
   between the PR's branch point and now AND the new service_tier
   strip.  Behavioural intent of both is preserved.

3. Six new tests in tests/agent/transports/test_codex_transport.py
   covering:
   - TestCodexTransportXaiServiceTierStrip (3 tests): xAI strips
     service_tier from request_overrides; non-xAI codex_responses
     and GitHub Models both KEEP service_tier (regression guards
     so the strip stays xAI-only).
   - TestPreflightSlashEnumStrip (3 tests): Grok and aggregator-
     prefixed Grok model names both trigger the safety-net strip;
     non-Grok models preserve slash enums as a regression guard
     against the strip becoming too broad.

51/51 in tests/agent/transports/test_codex_transport.py.

Co-authored-by: Nami4D <hello@nami4d.tech>
Bryce-huang pushed a commit to wbkunlun/hermes-agent that referenced this pull request May 29, 2026
@teknium1
fix(xai-oauth): gate slash-enum strip on model name + add regression …
54dd49d 
…tests (NousResearch#28490)

Three additions on top of @Nami4D's salvage:

1. Gate the preflight slash-enum strip on the model name pattern
   (grok-* / x-ai/grok-*).  The original PR stripped slash-containing
   enum values from every codex_responses request, but native Codex
   (OpenAI) and GitHub Models DO accept slash enums — stripping them
   there would silently degrade tool-schema constraints.  xAI is the
   only Responses-API surface that rejects the shape.

2. Resolve the merge conflict in agent/transports/codex.py by
   preserving both the timeout-forwarding block that landed on main
   between the PR's branch point and now AND the new service_tier
   strip.  Behavioural intent of both is preserved.

3. Six new tests in tests/agent/transports/test_codex_transport.py
   covering:
   - TestCodexTransportXaiServiceTierStrip (3 tests): xAI strips
     service_tier from request_overrides; non-xAI codex_responses
     and GitHub Models both KEEP service_tier (regression guards
     so the strip stays xAI-only).
   - TestPreflightSlashEnumStrip (3 tests): Grok and aggregator-
     prefixed Grok model names both trigger the safety-net strip;
     non-Grok models preserve slash enums as a regression guard
     against the strip becoming too broad.

51/51 in tests/agent/transports/test_codex_transport.py.

Co-authored-by: Nami4D <hello@nami4d.tech>

#AI commit#
mosaiq-systems pushed a commit to mosaiq-systems/hermes-agent that referenced this pull request May 29, 2026
@teknium1 @Nami4D
fix(xai-oauth): gate slash-enum strip on model name + add regression …
ef6d6f3 
…tests (NousResearch#28490)

Three additions on top of @Nami4D's salvage:

1. Gate the preflight slash-enum strip on the model name pattern
   (grok-* / x-ai/grok-*).  The original PR stripped slash-containing
   enum values from every codex_responses request, but native Codex
   (OpenAI) and GitHub Models DO accept slash enums — stripping them
   there would silently degrade tool-schema constraints.  xAI is the
   only Responses-API surface that rejects the shape.

2. Resolve the merge conflict in agent/transports/codex.py by
   preserving both the timeout-forwarding block that landed on main
   between the PR's branch point and now AND the new service_tier
   strip.  Behavioural intent of both is preserved.

3. Six new tests in tests/agent/transports/test_codex_transport.py
   covering:
   - TestCodexTransportXaiServiceTierStrip (3 tests): xAI strips
     service_tier from request_overrides; non-xAI codex_responses
     and GitHub Models both KEEP service_tier (regression guards
     so the strip stays xAI-only).
   - TestPreflightSlashEnumStrip (3 tests): Grok and aggregator-
     prefixed Grok model names both trigger the safety-net strip;
     non-Grok models preserve slash enums as a regression guard
     against the strip becoming too broad.

51/51 in tests/agent/transports/test_codex_transport.py.

Co-authored-by: Nami4D <hello@nami4d.tech>
gweeteve pushed a commit to gweeteve/hermes-agent that referenced this pull request Jun 2, 2026
@teknium1 @Nami4D
fix(xai-oauth): gate slash-enum strip on model name + add regression …
f8ad880 
…tests (NousResearch#28490)

Three additions on top of @Nami4D's salvage:

1. Gate the preflight slash-enum strip on the model name pattern
   (grok-* / x-ai/grok-*).  The original PR stripped slash-containing
   enum values from every codex_responses request, but native Codex
   (OpenAI) and GitHub Models DO accept slash enums — stripping them
   there would silently degrade tool-schema constraints.  xAI is the
   only Responses-API surface that rejects the shape.

2. Resolve the merge conflict in agent/transports/codex.py by
   preserving both the timeout-forwarding block that landed on main
   between the PR's branch point and now AND the new service_tier
   strip.  Behavioural intent of both is preserved.

3. Six new tests in tests/agent/transports/test_codex_transport.py
   covering:
   - TestCodexTransportXaiServiceTierStrip (3 tests): xAI strips
     service_tier from request_overrides; non-xAI codex_responses
     and GitHub Models both KEEP service_tier (regression guards
     so the strip stays xAI-only).
   - TestPreflightSlashEnumStrip (3 tests): Grok and aggregator-
     prefixed Grok model names both trigger the safety-net strip;
     non-Grok models preserve slash enums as a regression guard
     against the strip becoming too broad.

51/51 in tests/agent/transports/test_codex_transport.py.

Co-authored-by: Nami4D <hello@nami4d.tech>
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

comp/agent Core agent loop, run_agent.py, prompt builder P3 Low — cosmetic, nice to have provider/xai xAI (Grok) type/bug Something isn't working

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@Nami4D @teknium1 @alt-glitch