docs(anthropic): clarify OAuth uses Claude Pro/Max subscription usage by teknium1 · Pull Request #17399 · NousResearch/hermes-agent

teknium1 · 2026-04-29T10:58:52Z

Users have been asking what they're billed for when they authenticate Anthropic via OAuth in Hermes. This PR states it plainly in the docs: OAuth routes through Anthropic's Claude Code subscription path and consumes the extra Claude Code usage included with your Pro or Max plan; an ANTHROPIC_API_KEY is pay-per-token against that key's org instead.

Changes

integrations/providers.md: new :::info admonition in the Anthropic (Native) section + updated provider-table row.
user-guide/features/credential-pools.md: OAuth command comment line.
reference/environment-variables.md: Provider Auth (OAuth) intro paragraph.
getting-started/quickstart.md: provider-picker table row.

Validation

4 files changed, +10/-4. Docusaurus build output is unchanged apart from my edits (pre-existing missing-file error in src/pages/skills is unrelated).

Users have been asking what they're billed for when they authenticate Anthropic via OAuth in Hermes. Clarify in the provider docs that OAuth routes through Anthropic's Claude Code subscription path — consuming the extra Claude Code usage included with their Pro or Max plan — and that an ANTHROPIC_API_KEY is pay-per-token against that key's org instead. Touches: - integrations/providers.md: new info admonition in Anthropic (Native) section, plus provider-table row. - user-guide/features/credential-pools.md: OAuth comment line. - reference/environment-variables.md: Provider Auth (OAuth) intro. - getting-started/quickstart.md: provider-picker table row.

…s only (#17404) The previous docs pass (#17399) overstated what Anthropic OAuth works with. In practice Hermes can only route against a Claude Max plan that has purchased extra usage credits — the base Max allowance is not consumed, and Claude Pro is not supported at all. Without Max + extra credits, users must fall back to an ANTHROPIC_API_KEY (pay-per-token). Updates the four pages touched in #17399: - integrations/providers.md - user-guide/features/credential-pools.md - reference/environment-variables.md - getting-started/quickstart.md

) * docs(anthropic): correct OAuth scope to Max plan + extra usage credits only The previous docs pass (#17399) overstated what Anthropic OAuth works with. In practice Hermes can only route against a Claude Max plan that has purchased extra usage credits — the base Max allowance is not consumed, and Claude Pro is not supported at all. Without Max + extra credits, users must fall back to an ANTHROPIC_API_KEY (pay-per-token). Updates the four pages touched in #17399: - integrations/providers.md - user-guide/features/credential-pools.md - reference/environment-variables.md - getting-started/quickstart.md * fix(aux): skip kimi-coding in vision auto-detect (closes #17076) Kimi Coding Plan's /coding endpoint (Anthropic Messages wire) has no image_in capability — Kimi's own docs confirm and suggest switching to a vision-capable model. Vision lives on the separate Kimi Platform (api.moonshot.ai, OpenAI-wire, pay-as-you-go). When the user has kimi-coding as main provider and auxiliary.vision.provider=auto, resolve_vision_provider_client was handing back an AnthropicAuxiliaryClient wrapped around /coding which 404'd on every vision request. Add a _PROVIDERS_WITHOUT_VISION frozenset ({kimi-coding, kimi-coding-cn}) and gate the main-provider vision branch on membership. On a skip the auto-detect falls through to OpenRouter → Nous like any other main-provider-unavailable case. Explicit per-task overrides (auxiliary.vision.provider=kimi-coding) are unaffected — the skip only applies when the caller is in auto mode. Tests: 4 new targeted tests in TestVisionAutoSkipsKimiCoding covering the skip path, CN variant, explicit-override passthrough, and a guard against accidental skip-list widening.

…NousResearch#17399) Users have been asking what they're billed for when they authenticate Anthropic via OAuth in Hermes. Clarify in the provider docs that OAuth routes through Anthropic's Claude Code subscription path — consuming the extra Claude Code usage included with their Pro or Max plan — and that an ANTHROPIC_API_KEY is pay-per-token against that key's org instead. Touches: - integrations/providers.md: new info admonition in Anthropic (Native) section, plus provider-table row. - user-guide/features/credential-pools.md: OAuth comment line. - reference/environment-variables.md: Provider Auth (OAuth) intro. - getting-started/quickstart.md: provider-picker table row.

…s only (NousResearch#17404) The previous docs pass (NousResearch#17399) overstated what Anthropic OAuth works with. In practice Hermes can only route against a Claude Max plan that has purchased extra usage credits — the base Max allowance is not consumed, and Claude Pro is not supported at all. Without Max + extra credits, users must fall back to an ANTHROPIC_API_KEY (pay-per-token). Updates the four pages touched in NousResearch#17399: - integrations/providers.md - user-guide/features/credential-pools.md - reference/environment-variables.md - getting-started/quickstart.md

…#17076) (NousResearch#17451) * docs(anthropic): correct OAuth scope to Max plan + extra usage credits only The previous docs pass (NousResearch#17399) overstated what Anthropic OAuth works with. In practice Hermes can only route against a Claude Max plan that has purchased extra usage credits — the base Max allowance is not consumed, and Claude Pro is not supported at all. Without Max + extra credits, users must fall back to an ANTHROPIC_API_KEY (pay-per-token). Updates the four pages touched in NousResearch#17399: - integrations/providers.md - user-guide/features/credential-pools.md - reference/environment-variables.md - getting-started/quickstart.md * fix(aux): skip kimi-coding in vision auto-detect (closes NousResearch#17076) Kimi Coding Plan's /coding endpoint (Anthropic Messages wire) has no image_in capability — Kimi's own docs confirm and suggest switching to a vision-capable model. Vision lives on the separate Kimi Platform (api.moonshot.ai, OpenAI-wire, pay-as-you-go). When the user has kimi-coding as main provider and auxiliary.vision.provider=auto, resolve_vision_provider_client was handing back an AnthropicAuxiliaryClient wrapped around /coding which 404'd on every vision request. Add a _PROVIDERS_WITHOUT_VISION frozenset ({kimi-coding, kimi-coding-cn}) and gate the main-provider vision branch on membership. On a skip the auto-detect falls through to OpenRouter → Nous like any other main-provider-unavailable case. Explicit per-task overrides (auxiliary.vision.provider=kimi-coding) are unaffected — the skip only applies when the caller is in auto mode. Tests: 4 new targeted tests in TestVisionAutoSkipsKimiCoding covering the skip path, CN variant, explicit-override passthrough, and a guard against accidental skip-list widening.