mirror of
https://github.com/tinyhumansai/openhuman.git
synced 2026-07-28 13:32:23 +00:00
Steven Enamakel
@@ -1087,8 +1087,8 @@ pub(super) const CAPABILITIES: &[Capability] = &[
|
||||
name: "Configure AI",
|
||||
domain: "settings",
|
||||
category: CapabilityCategory::Settings,
|
||||
description: "Adjust AI-related settings and agent behavior preferences.",
|
||||
how_to: "Settings > Developer Options > AI Configuration",
|
||||
description: "Configure managed, local, custom, and built-in BYOK LLM providers, including SumoPod and other OpenAI-compatible gateways, plus per-workload routing preferences.",
|
||||
how_to: "Settings > AI",
|
||||
status: CapabilityStatus::Stable,
|
||||
privacy: None,
|
||||
},
|
||||
|
||||
@@ -12,6 +12,179 @@
|
||||
use schemars::JsonSchema;
|
||||
use serde::{Deserialize, Serialize};
|
||||
|
||||
#[derive(Debug, Clone, Copy, PartialEq, Eq)]
|
||||
pub struct BuiltinCloudProvider {
|
||||
pub slug: &'static str,
|
||||
pub label: &'static str,
|
||||
pub endpoint: &'static str,
|
||||
pub auth_style: AuthStyle,
|
||||
}
|
||||
|
||||
pub const BUILTIN_CLOUD_PROVIDERS: &[BuiltinCloudProvider] = &[
|
||||
BuiltinCloudProvider {
|
||||
slug: "openhuman",
|
||||
label: "OpenHuman",
|
||||
endpoint: "https://api.openhuman.ai/v1",
|
||||
auth_style: AuthStyle::OpenhumanJwt,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "openai",
|
||||
label: "OpenAI",
|
||||
endpoint: "https://api.openai.com/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "anthropic",
|
||||
label: "Anthropic",
|
||||
endpoint: "https://api.anthropic.com/v1",
|
||||
auth_style: AuthStyle::Anthropic,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "openrouter",
|
||||
label: "OpenRouter",
|
||||
endpoint: "https://openrouter.ai/api/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "orcarouter",
|
||||
label: "OrcaRouter",
|
||||
endpoint: "https://api.orcarouter.ai/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "gmi",
|
||||
label: "GMI",
|
||||
endpoint: "https://api.gmi-serving.com/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "fireworks",
|
||||
label: "Fireworks",
|
||||
endpoint: "https://api.fireworks.ai/inference/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "moonshot",
|
||||
label: "Kimi (Moonshot)",
|
||||
endpoint: "https://api.moonshot.ai/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "groq",
|
||||
label: "Groq",
|
||||
endpoint: "https://api.groq.com/openai/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "mistral",
|
||||
label: "Mistral",
|
||||
endpoint: "https://api.mistral.ai/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "deepseek",
|
||||
label: "DeepSeek",
|
||||
endpoint: "https://api.deepseek.com/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "together",
|
||||
label: "Together AI",
|
||||
endpoint: "https://api.together.xyz/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "google",
|
||||
label: "Google Gemini",
|
||||
endpoint: "https://generativelanguage.googleapis.com/v1beta/openai",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "cerebras",
|
||||
label: "Cerebras",
|
||||
endpoint: "https://api.cerebras.ai/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "xai",
|
||||
label: "xAI",
|
||||
endpoint: "https://api.x.ai/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "huggingface",
|
||||
label: "Hugging Face",
|
||||
endpoint: "https://router.huggingface.co/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "nvidia",
|
||||
label: "NVIDIA",
|
||||
endpoint: "https://integrate.api.nvidia.com/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "zai",
|
||||
label: "Z.AI",
|
||||
endpoint: "https://api.z.ai/api/paas/v4",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "minimax",
|
||||
label: "MiniMax",
|
||||
endpoint: "https://api.minimax.io/anthropic",
|
||||
auth_style: AuthStyle::Anthropic,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "stepfun",
|
||||
label: "StepFun",
|
||||
endpoint: "https://api.stepfun.ai/step_plan/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "kilocode",
|
||||
label: "Kilo Code",
|
||||
endpoint: "https://api.kilo.ai/api/gateway",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "deepinfra",
|
||||
label: "DeepInfra",
|
||||
endpoint: "https://api.deepinfra.com/v1/openai",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "novita",
|
||||
label: "Novita",
|
||||
endpoint: "https://api.novita.ai/v3/openai",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "venice",
|
||||
label: "Venice",
|
||||
endpoint: "https://api.venice.ai/api/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "vercel-ai-gateway",
|
||||
label: "Vercel AI Gateway",
|
||||
endpoint: "https://ai-gateway.vercel.sh/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
BuiltinCloudProvider {
|
||||
slug: "sumopod",
|
||||
label: "SumoPod",
|
||||
endpoint: "https://ai.sumopod.com/v1",
|
||||
auth_style: AuthStyle::Bearer,
|
||||
},
|
||||
];
|
||||
|
||||
fn builtin_cloud_provider(type_str: &str) -> Option<&'static BuiltinCloudProvider> {
|
||||
BUILTIN_CLOUD_PROVIDERS
|
||||
.iter()
|
||||
.find(|provider| provider.slug == type_str)
|
||||
}
|
||||
|
||||
/// Authentication header style for a cloud provider.
|
||||
///
|
||||
/// Wire format is lowercase (e.g. `"bearer"`). Determines which HTTP headers
|
||||
@@ -160,41 +333,24 @@ pub fn migrate_legacy_fields(entry: &mut CloudProviderCreds) {
|
||||
|
||||
// Auth style from legacy type when still at default Bearer.
|
||||
if entry.auth_style == AuthStyle::Bearer {
|
||||
match lt {
|
||||
"anthropic" => {
|
||||
entry.auth_style = AuthStyle::Anthropic;
|
||||
}
|
||||
"openhuman" => {
|
||||
entry.auth_style = AuthStyle::OpenhumanJwt;
|
||||
}
|
||||
_ => {}
|
||||
if let Some(provider) = builtin_cloud_provider(lt) {
|
||||
entry.auth_style = provider.auth_style;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/// Map a legacy type string (or slug) to a human-readable label.
|
||||
fn legacy_label_for(type_str: &str) -> &'static str {
|
||||
match type_str {
|
||||
"openhuman" => "OpenHuman",
|
||||
"openai" => "OpenAI",
|
||||
"anthropic" => "Anthropic",
|
||||
"openrouter" => "OpenRouter",
|
||||
"orcarouter" => "OrcaRouter",
|
||||
"custom" => "Custom",
|
||||
_ => "Custom",
|
||||
}
|
||||
builtin_cloud_provider(type_str)
|
||||
.map(|provider| provider.label)
|
||||
.unwrap_or("Custom")
|
||||
}
|
||||
|
||||
/// Map a legacy type string to its well-known default endpoint.
|
||||
fn legacy_default_endpoint(type_str: &str) -> &'static str {
|
||||
match type_str {
|
||||
"openhuman" => "https://api.openhuman.ai/v1",
|
||||
"openai" => "https://api.openai.com/v1",
|
||||
"anthropic" => "https://api.anthropic.com/v1",
|
||||
"openrouter" => "https://openrouter.ai/api/v1",
|
||||
"orcarouter" => "https://api.orcarouter.ai/v1",
|
||||
_ => "",
|
||||
}
|
||||
builtin_cloud_provider(type_str)
|
||||
.map(|provider| provider.endpoint)
|
||||
.unwrap_or("")
|
||||
}
|
||||
|
||||
/// Generate a short opaque id for a new provider entry.
|
||||
@@ -303,7 +459,10 @@ impl CloudProviderType {
|
||||
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
use super::is_slug_reserved;
|
||||
use super::{
|
||||
is_slug_reserved, migrate_legacy_fields, AuthStyle, CloudProviderCreds,
|
||||
BUILTIN_CLOUD_PROVIDERS,
|
||||
};
|
||||
|
||||
#[test]
|
||||
fn reserved_slugs() {
|
||||
@@ -329,4 +488,58 @@ mod tests {
|
||||
"lmstudio is a free-form OpenAI-compatible slug"
|
||||
);
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn builtin_cloud_provider_defaults_cover_phase_one_presets() {
|
||||
for (slug, label, endpoint, auth_style) in [
|
||||
(
|
||||
"groq",
|
||||
"Groq",
|
||||
"https://api.groq.com/openai/v1",
|
||||
AuthStyle::Bearer,
|
||||
),
|
||||
(
|
||||
"deepseek",
|
||||
"DeepSeek",
|
||||
"https://api.deepseek.com/v1",
|
||||
AuthStyle::Bearer,
|
||||
),
|
||||
(
|
||||
"minimax",
|
||||
"MiniMax",
|
||||
"https://api.minimax.io/anthropic",
|
||||
AuthStyle::Anthropic,
|
||||
),
|
||||
(
|
||||
"sumopod",
|
||||
"SumoPod",
|
||||
"https://ai.sumopod.com/v1",
|
||||
AuthStyle::Bearer,
|
||||
),
|
||||
] {
|
||||
let mut entry = CloudProviderCreds {
|
||||
id: format!("p_{slug}"),
|
||||
legacy_type: Some(slug.to_string()),
|
||||
..Default::default()
|
||||
};
|
||||
migrate_legacy_fields(&mut entry);
|
||||
|
||||
assert_eq!(entry.slug, slug);
|
||||
assert_eq!(entry.label, label);
|
||||
assert_eq!(entry.endpoint, endpoint);
|
||||
assert_eq!(entry.auth_style, auth_style);
|
||||
}
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn builtin_cloud_provider_slugs_are_unique() {
|
||||
let mut slugs = std::collections::HashSet::new();
|
||||
for provider in BUILTIN_CLOUD_PROVIDERS {
|
||||
assert!(
|
||||
slugs.insert(provider.slug),
|
||||
"duplicate built-in cloud provider slug {}",
|
||||
provider.slug
|
||||
);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@@ -12,46 +12,49 @@ Unified inference domain: the canonical home for everything LLM/STT/TTS/embeddin
|
||||
- Run ChatGPT/Codex OAuth (PKCE) for the `openai` cloud slug and persist tokens in the encrypted auth-profile store.
|
||||
- Expose an OpenAI-compatible `/v1/*` HTTP endpoint guarded by a stable user-managed external bearer.
|
||||
- Detect device hardware profile and recommend/apply local model presets/tiers.
|
||||
- Maintain the built-in BYOK provider preset catalog used by Settings > AI.
|
||||
The current matrix lives in `docs/inference-provider-catalog.md`; credentials
|
||||
are stored under `provider:<slug>` in the auth-profile store.
|
||||
|
||||
## Key files
|
||||
|
||||
| File / dir | Role |
|
||||
| --- | --- |
|
||||
| `mod.rs` | Domain root; module decls + re-exports; wires `inference.*` controller schemas/controllers. |
|
||||
| `ops.rs` | Canonical handler file — `inference_*` business logic returning `RpcOutcome<T>`; delegates to `local`, `provider`, `sentiment`, `device`, `presets`, `openai_oauth`. Includes Sentry-noise suppression for expected provider/user-config failures. |
|
||||
| `schemas.rs` | `inference.*` controller schemas + `handle_*` fns + param DTOs. |
|
||||
| `types.rs` | Serde DTOs: `LocalAiStatus`, `LocalAiAssetsStatus`, `LocalAiDownloadsProgress`, `LocalAiEmbeddingResult`, `LocalAiSpeechResult`, `LocalAiTtsResult`, etc. |
|
||||
| `device.rs` | `DeviceProfile` hardware detection (RAM/CPU/GPU/OS), cached. |
|
||||
| `model_ids.rs` | Effective chat/vision/embedding/STT/TTS/quantization model id resolution from config. |
|
||||
| `model_context.rs` | Known model context-window sizes (`context_window_for_model`) for pre-dispatch budgeting. |
|
||||
| `presets.rs` | `ModelPreset`, `ModelTier`, `VisionMode`; tier recommendation + apply-to-config; MVP preset gating. |
|
||||
| `sentiment.rs` | `SentimentResult` + emotion/valence analysis via the local model. |
|
||||
| `parse.rs` / `paths.rs` | Output parsing helpers / on-disk model artifact paths. |
|
||||
| `local/` | Local runtime manager (was `local_ai/`). |
|
||||
| `local/core.rs` | `LocalAiService` singleton (`global`/`try_global`), `model_artifact_path`. |
|
||||
| `local/ops.rs` | Local RPC entrypoints (`local_ai_status/prompt/summarize/vision_prompt/embed/should_react`, `ReactionDecision`); re-exported as `local::rpc`. |
|
||||
| `local/schemas.rs` | `local_ai.*` controller schemas + handlers. |
|
||||
| `local/ollama.rs`, `local/lm_studio.rs` | Provider-specific runtime drivers; base-url resolution. |
|
||||
| `local/install*.rs`, `local/voice_install_common.rs` | Whisper/Piper install + shared download logic. |
|
||||
| `local/model_requirements.rs` | `MIN_CONTEXT_TOKENS`, `evaluate_context`, `ContextEligibility`. |
|
||||
| `local/service/` | `LocalAiService` impl split: `bootstrap`, `ollama_admin`, `public_infer`, `speech`, `vision_embed`, `whisper_engine`, `assets`, `spawn_marker`. |
|
||||
| `provider/` | Unified provider abstraction (was `providers/`). |
|
||||
| `provider/traits.rs` | `Provider` trait + `ChatMessage`/`ChatRequest`/`ChatResponse`/`ToolCall`/`UsageInfo`/`ProviderDelta` etc. |
|
||||
| `provider/factory.rs` | `create_chat_provider`, `provider_for_role`, provider-string grammar, local/cloud construction; `BYOK_INCOMPLETE_SENTINEL`. |
|
||||
| `provider/router.rs` | `RouterProvider` hint-based multi-model routing. |
|
||||
| `provider/reliable.rs` | Retry/backoff wrapper. |
|
||||
| `provider/compatible*.rs` | OpenAI-compatible provider (request dump/parse/stream/types). |
|
||||
| `provider/openhuman_backend.rs` | Managed OpenHuman backend provider (session JWT). |
|
||||
| `provider/claude_agent_sdk/` | Claude Agent SDK subprocess provider (`protocol.rs`, `subprocess.rs`). |
|
||||
| `provider/config_rejection.rs`, `provider/billing_error.rs` | Error classifiers (unknown-model / config rejection / budget exhausted). |
|
||||
| `provider/temperature.rs`, `provider/thread_context.rs` | Per-workload temperature override; thread context plumbing. |
|
||||
| `provider/ops.rs` | `list_configured_models`, SessionExpired publishing on auth failure. |
|
||||
| `provider/schemas.rs` | Provider-layer schemas. |
|
||||
| `voice/` | Inference implementations imported by `crate::openhuman::voice`. |
|
||||
| `voice/cloud_transcribe.rs`, `voice/local_transcribe.rs`, `voice/local_speech.rs` | STT (cloud + local) and local TTS. |
|
||||
| `voice/streaming.rs`, `voice/postprocess.rs`, `voice/hallucination.rs` | Streaming transcription, post-processing, hallucination filtering. |
|
||||
| `openai_oauth/` | ChatGPT/Codex OAuth: `config.rs` (Codex OAuth config), `flow.rs` (start/complete/status/disconnect), `store.rs` (token persistence). |
|
||||
| `http/` | OpenAI-compatible endpoint: `server.rs` (`router()`), `types.rs`; `EXTERNAL_OPENAI_COMPAT_PROVIDER` bearer id. |
|
||||
| File / dir | Role |
|
||||
| --------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
||||
| `mod.rs` | Domain root; module decls + re-exports; wires `inference.*` controller schemas/controllers. |
|
||||
| `ops.rs` | Canonical handler file — `inference_*` business logic returning `RpcOutcome<T>`; delegates to `local`, `provider`, `sentiment`, `device`, `presets`, `openai_oauth`. Includes Sentry-noise suppression for expected provider/user-config failures. |
|
||||
| `schemas.rs` | `inference.*` controller schemas + `handle_*` fns + param DTOs. |
|
||||
| `types.rs` | Serde DTOs: `LocalAiStatus`, `LocalAiAssetsStatus`, `LocalAiDownloadsProgress`, `LocalAiEmbeddingResult`, `LocalAiSpeechResult`, `LocalAiTtsResult`, etc. |
|
||||
| `device.rs` | `DeviceProfile` hardware detection (RAM/CPU/GPU/OS), cached. |
|
||||
| `model_ids.rs` | Effective chat/vision/embedding/STT/TTS/quantization model id resolution from config. |
|
||||
| `model_context.rs` | Known model context-window sizes (`context_window_for_model`) for pre-dispatch budgeting. |
|
||||
| `presets.rs` | `ModelPreset`, `ModelTier`, `VisionMode`; tier recommendation + apply-to-config; MVP preset gating. |
|
||||
| `sentiment.rs` | `SentimentResult` + emotion/valence analysis via the local model. |
|
||||
| `parse.rs` / `paths.rs` | Output parsing helpers / on-disk model artifact paths. |
|
||||
| `local/` | Local runtime manager (was `local_ai/`). |
|
||||
| `local/core.rs` | `LocalAiService` singleton (`global`/`try_global`), `model_artifact_path`. |
|
||||
| `local/ops.rs` | Local RPC entrypoints (`local_ai_status/prompt/summarize/vision_prompt/embed/should_react`, `ReactionDecision`); re-exported as `local::rpc`. |
|
||||
| `local/schemas.rs` | `local_ai.*` controller schemas + handlers. |
|
||||
| `local/ollama.rs`, `local/lm_studio.rs` | Provider-specific runtime drivers; base-url resolution. |
|
||||
| `local/install*.rs`, `local/voice_install_common.rs` | Whisper/Piper install + shared download logic. |
|
||||
| `local/model_requirements.rs` | `MIN_CONTEXT_TOKENS`, `evaluate_context`, `ContextEligibility`. |
|
||||
| `local/service/` | `LocalAiService` impl split: `bootstrap`, `ollama_admin`, `public_infer`, `speech`, `vision_embed`, `whisper_engine`, `assets`, `spawn_marker`. |
|
||||
| `provider/` | Unified provider abstraction (was `providers/`). |
|
||||
| `provider/traits.rs` | `Provider` trait + `ChatMessage`/`ChatRequest`/`ChatResponse`/`ToolCall`/`UsageInfo`/`ProviderDelta` etc. |
|
||||
| `provider/factory.rs` | `create_chat_provider`, `provider_for_role`, provider-string grammar, local/cloud construction; `BYOK_INCOMPLETE_SENTINEL`. |
|
||||
| `provider/router.rs` | `RouterProvider` hint-based multi-model routing. |
|
||||
| `provider/reliable.rs` | Retry/backoff wrapper. |
|
||||
| `provider/compatible*.rs` | OpenAI-compatible provider (request dump/parse/stream/types). |
|
||||
| `provider/openhuman_backend.rs` | Managed OpenHuman backend provider (session JWT). |
|
||||
| `provider/claude_agent_sdk/` | Claude Agent SDK subprocess provider (`protocol.rs`, `subprocess.rs`). |
|
||||
| `provider/config_rejection.rs`, `provider/billing_error.rs` | Error classifiers (unknown-model / config rejection / budget exhausted). |
|
||||
| `provider/temperature.rs`, `provider/thread_context.rs` | Per-workload temperature override; thread context plumbing. |
|
||||
| `provider/ops.rs` | `list_configured_models`, SessionExpired publishing on auth failure. |
|
||||
| `provider/schemas.rs` | Provider-layer schemas. |
|
||||
| `voice/` | Inference implementations imported by `crate::openhuman::voice`. |
|
||||
| `voice/cloud_transcribe.rs`, `voice/local_transcribe.rs`, `voice/local_speech.rs` | STT (cloud + local) and local TTS. |
|
||||
| `voice/streaming.rs`, `voice/postprocess.rs`, `voice/hallucination.rs` | Streaming transcription, post-processing, hallucination filtering. |
|
||||
| `openai_oauth/` | ChatGPT/Codex OAuth: `config.rs` (Codex OAuth config), `flow.rs` (start/complete/status/disconnect), `store.rs` (token persistence). |
|
||||
| `http/` | OpenAI-compatible endpoint: `server.rs` (`router()`), `types.rs`; `EXTERNAL_OPENAI_COMPAT_PROVIDER` bearer id. |
|
||||
|
||||
## Public surface
|
||||
|
||||
|
||||
@@ -225,6 +225,7 @@ fn subagent_lifecycle_records_and_clears_active() {
|
||||
dedicated_thread: false,
|
||||
prompt_chars: 42,
|
||||
worker_thread_id: None,
|
||||
display_name: None,
|
||||
});
|
||||
let s = m.snapshot();
|
||||
assert_eq!(s.active_subagent.as_deref(), Some("researcher"));
|
||||
|
||||
Reference in New Issue
Block a user