feat: planner-aware profile routing, execution timeline metadata, and Discord /chat model switching (with Ollama support)

[jbax1899]

Delivers end-to-end model-profile routing and visibility improvements across backend and Discord.
Adds structured execution timeline metadata (planner/tool/generation with status/duration).
Introduces /api/chat/profiles and slash /chat per-request profile switching.
Supports multi-guild command rollout and profile-choice fetch retries.
Extends provider config/runtime wiring for Ollama (including cloud/local handling and profile enablement rules).

This branch intentionally grouped the model-profile routing seam with /chat operator controls and execution visibility so we could validate the full request path (selection -> resolution -> execution -> surfaced metadata). This also helps close integration gaps discovered during testing (profile override precedence, provenance/footer parity, and Ollama cloud runtime wiring) before splitting follow-up hardening into later branches.

Summary by CodeRabbit

• New Features

  • Added /chat Discord slash command with dynamic profile choices and generation hints (reasoning effort, verbosity).
  • Public API endpoint to list available chat profiles.
  • Multi-guild Discord command deployment.
  • Optional Ollama provider support (local and cloud).

• Documentation

  • Updated environment examples to document Ollama settings and multi-guild deployment.

• Bug Fixes & Improvements

  • Enhanced execution/timing metadata, provenance display, and more robust profile/model selection.

[coderabbitai]

Caution

Review failed

The pull request is closed.

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 2559f07c-0883-4716-8b64-53aa994de266

📥 Commits

Reviewing files that changed from the base of the PR and between 9fe9abb and b924898.

📒 Files selected for processing (1)

• packages/backend/src/handlers/chatProfiles.ts

---

📝 Walkthrough

Walkthrough

Adds Ollama provider support, a GET /api/chat/profiles endpoint and client, execution-timeline metadata (planner/tool/generation), planner/orchestrator telemetry and fail-open behavior, multi-guild Discord slash-command deployment with a new /chat command and profile-choice hydration, plus related config, contracts, runtime, UI, and tests.

Changes

Cohort / File(s)	Summary
Env & Config Examples \.env\.example, packages/config-spec/src/env-spec.ts	Add OLLAMA_BASE_URL, OLLAMA_API_KEY, OLLAMA_LOCAL_INFERENCE_ENABLED; mark OPENAI_API_KEY optional; update env metadata.
Backend Config Sections & Parsers packages/backend/src/config/sections/ollama.ts, packages/backend/src/config/buildRuntimeConfig.ts, packages/backend/src/config/types.ts, packages/backend/src/config/parsers.ts	Add Ollama section builder, wire Ollama into runtime config/types, add parseBooleanFlag, and include Ollama env values in startup diagnostics.
Model Profiles Catalog packages/backend/src/config/model-profiles.defaults.yaml, packages/backend/src/config/sections/modelProfiles.ts	Add Ollama model profiles; compute provider availability from env and disable profiles whose provider is unavailable (emit warnings).
API Spec & Client docs/api/openapi.yaml, packages/api-client/src/chat.ts	Add GET /api/chat/profiles endpoint spec and client method getChatProfiles with response validation.
Contracts & Schemas packages/contracts/src/web/types.ts, packages/contracts/src/web/schemas.ts, packages/contracts/src/ethics-core/types.ts, packages/contracts/src/ethics-core/index.ts, packages/contracts/src/index.ts, packages/contracts/src/providers.ts	Add ChatProfileOption/GetChatProfilesResponse, add generation to PostChatRequest, add execution-telemetry types/schemas (ExecutionEvent, ExecutionStatus, ExecutionReasonCode), extend ResponseMetadata with totalDurationMs and execution[], and include ollama in supported providers.
Chat Profiles Handler & Routing packages/backend/src/handlers/chatProfiles.ts, packages/backend/src/server.ts	New GET /api/chat/profiles handler returning enabled profiles (trim/truncate descriptions), method-guard (405), fail-open fallback, and server wiring.
Agent Runtime (VoltAgent) packages/agent-runtime/src/voltagentRuntime.ts, packages/agent-runtime/test/voltagentRuntime.test.ts	Generalize provider-tool shape, add Ollama executor config and provider remapping (ollama → ollama-cloud when applicable), provider-based search-tool registry, and corresponding tests.
Planner & Orchestrator packages/backend/src/services/chatPlanner.ts, packages/backend/src/services/chatOrchestrator.ts, packages/backend/src/services/chatProfileOverlay.ts	Planner now returns execution metadata (status/reason/duration); orchestrator records timings, changes profile-resolution precedence and merges request generation overrides; overlay no longer warns on profile mismatch.
Chat Service & Metadata packages/backend/src/services/chatService.ts, packages/backend/src/services/openaiService.ts	Measure generation duration, accept orchestration/execution context, compute/propagate totalDurationMs, and build canonical execution timeline into response metadata.
Discord Bot: Command & Wiring packages/discord-bot/src/commands/chat.ts, packages/discord-bot/src/utils/chatCommandProfiles.ts, packages/discord-bot/src/config/runtime.ts, packages/discord-bot/src/index.ts, packages/discord-bot/src/utils/MessageProcessor.ts	Add /chat slash command (profile/reasoning/verbosity), apply backend profile choices (retries, truncate to 25), switch to DISCORD_GUILD_IDS multi-guild deployment with deduplication and per-guild deployment handling, and remove hardcoded profileId from message processor payload.
UI: Trace & Provenance packages/web/src/components/ProvenanceFooter.tsx, packages/web/src/pages/TracePage.tsx	Render execution timeline summary and totalDurationMs in provenance/trace UI and derive model label from execution when present.
Tests packages/backend/test/*, packages/discord-bot/test/*, packages/agent-runtime/test/*, packages/contracts/test/*	Add/adjust tests for Ollama routing, chat profiles endpoint, execution telemetry, planner failure handling, Discord /chat command behavior and profile-choice hydration, and schema validation for execution events.

Sequence Diagram(s)

sequenceDiagram
    participant User as Discord User
    participant Bot as Discord Bot
    participant Backend as Backend API
    participant Planner as Chat Planner
    participant Generation as Generation Runtime

    User->>Bot: /chat (prompt, optional profile/generation)
    Bot->>Bot: Defer reply
    Bot->>Backend: POST /api/chat (with orchestrationStartedAtMs & generation hint)
    activate Backend
    Backend->>Planner: planChat(request)
    activate Planner
    Planner-->>Backend: { plan, execution: { status, durationMs } }
    deactivate Planner
    Backend->>Backend: Resolve profile (request override or planner/default)
    Backend->>Generation: generate(request with effective profile/model)
    activate Generation
    Generation-->>Backend: message + generation metadata (model,duration)
    deactivate Generation
    Backend->>Backend: Build execution timeline (planner → tool? → generation)
    Backend-->>Bot: { action: 'message', message, metadata (execution,totalDurationMs) }
    deactivate Backend
    Bot->>User: Edit reply with message and execution summary

Loading

sequenceDiagram
    participant Env as Env
    participant Config as Runtime Config
    participant Catalog as Model Catalog Parser

    Env->>Config: buildRuntimeConfig(env)
    activate Config
    Config->>Catalog: parse default model-profiles
    activate Catalog
    Catalog-->>Config: parsedCatalog
    deactivate Catalog
    Config->>Config: compute provider availability (OPENAI_API_KEY, OLLAMA_BASE_URL, OLLAMA_LOCAL_INFERENCE_ENABLED)
    Config->>Config: disable catalog entries whose provider is unavailable
    Config-->>Env: runtimeConfig with filtered modelProfiles
    deactivate Config

Loading

Estimated code review effort

🎯 4 (Complex) | ⏱️ ~60 minutes

Possibly related PRs

• Complete text migration off legacy OpenAI for reflect and /news #188: overlaps agent-runtime changes (voltagentRuntime provider/tool generalization).
• Enhance model profile handling in backend configuration and runtime. #197: related backend model-profile/catalog and availability resolution changes.
• feat: planner-aware profile routing, execution timeline metadata, and Discord /chat model switching (with Ollama support) #200: broadly overlapping cross-cutting changes (Ollama provider, chat profiles endpoint, execution telemetry, Discord multi-guild).

Suggested labels

enhancement

Poem

🐰 I sniffed the Ollama breeze,
Profiles lined up like springtime trees,
Planner ticks, the runtimes play,
Commands hop out to guilds today,
Traces gleam — carrots all the way! 🥕

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The PR title comprehensively summarizes the main changes: planner-aware profile routing (request-level profile override and planner integration), execution timeline metadata (new ExecutionEvent tracking), and Discord /chat command with model switching and Ollama provider support.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ 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 feat/second-provider-and-tool-registry

---

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 and usage tips.}