feat(provider): add RunInfra support

[zerob13]

RunInfra users currently need a custom provider profile. This PR adds a built-in runinfra profile with the official workspace API endpoint and logo.

• Reuse AiSdkProvider with the OpenAI-compatible Chat Completions transport at https://api.runinfra.ai/v1; no new adapter or dependency.
• Discover models and check credentials through authenticated GET /v1/models, without generating billable inference. API keys use existing main-process provider storage.
• Add the static official SVG mark with the existing monochrome light/dark theme handling.
• Cover default-profile model discovery, Bearer authentication, and invalid-key errors with a focused contract test.

Validation passed:

• pnpm run format and pnpm run format:check
• pnpm run i18n
• pnpm run lint
• pnpm run typecheck
• 100 main tests across provider defaults, registration, runtime, factory, reasoning wire, and OpenAI compatibility.
• 37 renderer tests covering provider settings, API configuration, and model icons.
• SVG checked at small sizes in light and dark themes.

BEFORE
Provider catalog -> RunInfra unavailable

AFTER
Provider catalog -> [R] RunInfra
                         |
                         v
Provider settings -> API Key [********]  [Refresh models]

API contract: RunInfra quickstart and model discovery.

Summary by CodeRabbit

• New Features

  • Added RunInfra as a supported AI provider.
  • Added RunInfra model discovery through its models endpoint.
  • Added API key authentication and provider connection validation.
  • Added RunInfra branding and icon support.

• Tests

  • Added coverage for model discovery, successful authentication, and invalid API key handling.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: cc04e005-c3f1-41c7-bb00-499d69ce545b

📥 Commits

Reviewing files that changed from the base of the PR and between 8dada4b and fffbfa0.

⛔ Files ignored due to path filters (1)

• src/renderer/src/assets/llm-icons/runinfra.svg is excluded by !**/*.svg

📒 Files selected for processing (4)

• src/main/provider/defaults.ts
• src/main/provider/providerRegistry.ts
• src/renderer/src/components/icons/modelIconRegistry.ts
• test/main/provider/basicApiKeyProviders.test.ts

Included review availability: Your plan provides up to 8 included reviews per hour; 6 remain after this review.

---

📝 Walkthrough

Walkthrough

RunInfra is added as a disabled OpenAI-compatible provider with API-key authentication and no embedding support. Its model icon is registered. Tests cover model discovery and API-key validation.

Changes

RunInfra provider integration

Layer / File(s)	Summary
Register RunInfra provider src/main/provider/defaults.ts, src/main/provider/providerRegistry.ts, src/renderer/src/components/icons/modelIconRegistry.ts	Adds RunInfra defaults, provider registry settings, and monochrome model icon resolution.
Validate RunInfra API behavior test/main/provider/basicApiKeyProviders.test.ts	Tests provider resolution, model discovery, valid API keys, invalid API keys, and SDK call behavior.

Estimated code review effort: 2 (Simple) | ~10 minutes

Merge Risk: ⚪ Minimal · up to fffbf

This PR adds a localized built-in RunInfra provider profile, model discovery, authentication handling, and themed branding with validation in place. No actionable merge-blocking risk remains.

Sequence Diagram(s)

sequenceDiagram
  participant ProviderTest
  participant ProviderRegistry
  participant RunInfraAPI
  ProviderTest->>ProviderRegistry: resolve runinfra
  ProviderTest->>RunInfraAPI: request /v1/models with API key
  RunInfraAPI-->>ProviderTest: return models or 401
  ProviderTest->>ProviderRegistry: check API key result

Loading

Suggested reviewers: cuihuan

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly and concisely describes the main change: adding RunInfra provider support.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 4…
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

Full details: Docstring Coverage

Explanation

No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 4 files.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch codex/add-runinfra-provider

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[zerob13]

Review verdict: APPROVED (comment — self-PR cannot be formally approved)

Reviewed head fffbfa01a14dc581751498e22a6d634fbc8756ff (5 files, +87/−0, merge-base diff vs dev).

Checks

• Pattern conformance: exact template used by recently merged provider additions (e.g. #2215 Synthorai) — defaults.ts entry (disabled built-in, openai-completions, baseUrl + websites metadata), providerRegistry.ts id-mapping (OPENAI_BASE + credentialStrategy: 'api-key', embeddingStrategy: 'none', consistent with 31 other no-embedding entries), icon asset + modelIconRegistry color/mono registration.
• RunInfra is already present in resources/model-db/providers.json (auto-refreshed catalog), so the built-in defaults entry and the public model DB now align — no schema drift.
• Tests: appropriate, not excessive — one focused test covering model discovery via the OpenAI-compatible endpoint, fetchModels result shape, check() success + 401 error path, and Authorization: Bearer header. Mirrors the existing per-provider test granularity.
• Verification on PR branch: basicApiKeyProviders.test.ts 21/21 passed; oxfmt --check and pnpm lint clean. typecheck failures are the pre-existing markstream-vue/markdownWorkerLifecycle.ts renderer errors on current dev, unrelated to this PR.
• No over-engineering, no breaking changes, no UI-visible behavior change (provider stays enable: false until user opts in).

Nothing to change. Ship it.