The qwen3-embedding:4b model defaults to 2560-d vectors. Both
Ollama (client.embed()) and OpenAI support a dimensions parameter
to request a specific output size. This change threads the configured
qdrant.vectorSize through the AI provider layer so the model returns
vectors matching the Qdrant collection dimensions.
- AiApi.embeddings() now accepts optional dimensions parameter
- Ollama provider: switched from client.embeddings() to client.embed()
- OpenAI provider: passes dimensions to embeddings.create()
- VectorStoreService.getEmbedding() passes env.qdrant.vectorSize
- Added unit tests for dimension mismatch detection, collection
creation, and search guards
Adds vector-based semantic search across all chat sessions using Qdrant.
When a ChatTurn finishes, its content is chunked, embedded, and upserted
to a Qdrant collection. A search API and UI components enable searching
at user, project, and session scope.
Phase 1 — Configuration & Dependencies
- Add port/apiKey to GadgetCodeConfig.qdrant type
- Uncomment and update qdrant section in YAML config example
- Add qdrant config passthrough in env.ts
- Add @qdrant/js-client-rest dependency
Phase 2 — AI Embedding API (@gadget/ai)
- Add IAiEmbeddingResponse interface and abstract embeddings() to AiApi
- Implement embeddings() in OllamaAiApi (client.embeddings)
- Implement embeddings() in OpenAiApi (client.embeddings.create)
- Export IAiEmbeddingResponse from package index
Phase 3 — Backend Vector Store Service
- Create VectorStoreService (ingestTurn, search, removeTurnPoints)
- Hook fire-and-forget ingest after turn.save() in drone-session
- Register VectorStoreService in service startup/shutdown
Phase 4 — Backend Search API
- Create POST /api/v1/search controller with userId enforcement
- Batch-hydrate results from MongoDB (user, project, session, turn)
- Register search route in v1 API router
Phase 5 — Frontend Search Components
- SearchInput: debounced input with lucide-react icons
- ChatSearchResults: modal with score badges, metadata, loading states
- DroneSelectionModal: drone picker for sessions without a drone
- Add searchApi and ISearchResult to API client
- Add search to Home (global), ProjectManager (project), ChatSessionView (session)
- Add id=turn-{turnId} to ChatTurn for scroll targeting
- Scroll-to-turn from search result selection and router state
- Show DroneSelectionModal when no drone available
- Add Select Drone button in ChatSessionView sidebar
- Log the SDK response object immediately after client.chat.completions.create()
in both generate() and readStreamingChatCompletion()
- These debug-level logs capture the response at the call site before any
iteration or processing
- All downstream info-level logs continue to dump raw chunk/response objects
- Add rawUsage field to OpenAiChatIterationResult to carry the
original OpenAI SDK usage object through the call chain
- All downstream chat() logs now dump rawUsage via
JSON.parse(JSON.stringify()) instead of our cooked
{ promptTokens, completionTokens }
- This exposes completion_tokens_details.reasoning_tokens and
any other fields OpenAI returns that we weren't capturing
Phase 1: OpenAI API Token Extraction
- Add stream_options: { include_usage: true } to all streaming API calls
- Capture chunk.usage from final streaming chunks and response.usage from non-streaming
- Extend OpenAiChatIterationResult with optional usage field
- Update buildStats() to accept and return real token counts from usage data
- Wire iteration.usage through chat() to both buildStats() call sites
Phase 2: Agent Loop Stats Propagation
- Rename inputTokens/outputTokens to masterInputTokens/masterOutputTokens, add masterThinkingTokens
- Accumulate response.stats.tokenCounts after each master AI call
- Delete all Math.ceil(length/4) crude approximations (master and subagent loops)
- Track startTime/durationMs and emit IWorkOrderCompleteStats with workOrderComplete
- Subagent loop uses response.stats?.tokenCounts instead of Math.ceil
Phase 3: Database Model Changes
- Add contextWindowUsage field to IChatSession, ChatSessionSchema, and frontend ChatSession
- Initialize contextWindowUsage: 0 on session creation
Phase 4: Persist Stats on Turn Completion
- drone-session: accept IWorkOrderCompleteStats, persist turn stats, walk subagent records for aggregate
- drone-session: $inc session stats and contextWindowUsage, add formatDurationLabel() helper
- code-session: accept and forward stats, update in-memory session stats
- message-queue: Redis replay handles 4th stats arg
- Update WorkOrderCompleteMessage type in @gadget/api to accept stats parameter
Phase 5: UI — Context Window Fuel Gauge
- Add contextWindowUsage prop to SessionPanel
- Add fuel gauge bar with E→F labels, green/yellow/red zones, token count display
- Visible for ALL provider types (not gated by apiType)
Phase 6: Frontend Streaming State
- Add IWorkOrderCompleteStats interface to frontend api.ts
- handleWorkOrderComplete accepts stats, updates turn stats and session contextWindowUsage
- Pass contextWindowUsage prop to SessionPanel
New chat tool: chat_export
- Export current ChatSession and all finished ChatTurn records to disk
- Two formats: markdown (human-readable) and json (machine-readable)
- JSON exports always include a companion -readme.md with author credits,
Gadget Code version/link, AI provider/model/parameters
- Output to .gadget/exports/ with sanitized session name + timestamp
- Security-first: excludes emails, API keys, baseUrls, system prompts,
gitUrls, and other PII/infrastructure details from all exports
- Only finished/aborted/error turns included (processing excluded)
- Follows SubagentTool setter-injection pattern for session data
- Available in all chat session modes (plan/build/test/ship/dev)
Files:
- packages/ai-toolbox/src/chat/export.ts (NEW - 729 lines)
- packages/ai-toolbox/src/chat/index.ts (barrel update)
- gadget-drone/src/tools/index.ts (re-export update)
- gadget-drone/src/services/agent.ts (import, register, getExportData)
- Add detectWorkspace() utility in packages/api/src/lib/workspace-detector.ts
that walks up from process.cwd() looking for .gadget/workspace.json
- Update gadget-drone to detect workspace at startup, change to workspace
directory, and restore original startup directory on shutdown
- Update gadget-tasks with same workspace detection pattern
- Both tools now fail fast if started outside a managed workspace
- Prevents creating invalid workspaces when started from subdirectories
Plan tools extended to all modes; recent chat sessions added to
Authenticated Home view; drone locking and chat session startup factored
out of Project Manager into Chat Session view and made universal; update
for AGENTS.md.
Converts gadget-tasks from a duplicated codebase with direct MongoDB access
to a headless IDE client that uses gadget-code's REST API and Socket.IO
protocol to submit and process task prompts through the existing pipeline.
Deleted (no longer needed):
- gadget-tasks/src/models/ (5 Mongoose models — duplicated from gadget-code)
- gadget-tasks/data/prompts/ (11 prompt templates — duplicated from gadget-code)
- gadget-tasks/src/services/executor.ts (629-line AWL loop — duplicated from gadget-drone)
- gadget-tasks/src/services/ai.ts (direct AI API calls — drones do this now)
- gadget-tasks/src/services/workspace.ts (workspace management — drones do this now)
Added:
- gadget-tasks/src/services/platform.ts — REST API + Socket.IO headless IDE client
with session lock, workspace mode, prompt submission, work order tracking
Rewritten:
- gadget-tasks/src/gadget-tasks.ts — new startup: user login → auth → drone
selection → Socket.IO connect → schedule tasks (no MongoDB)
- gadget-tasks/src/services/scheduler.ts — delegates to PlatformService.executeTask()
- gadget-tasks/src/config/env.ts — platform.baseUrl config, no mongodb/google
gadget-code changes:
- Added PATCH /api/v1/projects/:projectId/tasks/:taskId/lastRun route
- Added ProjectService.updateTaskLastRun() for atomic task field update
Config changes:
- GadgetTasksConfig: replaced mongodb with platform.baseUrl, removed google.cse
- Dependencies: removed mongoose, @gadget/ai, @gadget/ai-toolbox, dayjs,
simple-git, nanoid; added socket.io-client, @inquirer/prompts
- IChatSession/IChatTurn interfaces: add optional numCtx field
- Mongoose schemas: add numCtx field to ChatSession and ChatTurn
- API controller: validate numCtx (>=16384, multiple of 8192)
- ChatSessionService: persist numCtx on session, pass through to turn
- SessionPanel: add range slider for Context Window (hidden for OpenAI)
- Drone buildDroneModelConfig: prefer turn.numCtx over model defaults
- Create comprehensive plan document for agent instructions text area
- Define requirements, acceptance criteria, and technical implementation
- Include UI/UX mockups and testing strategy
- Plan discovered during FILES panel implementation
- Addresses need for project-specific acceptance criteria
Add end-to-end abort support: AbortSignal in @gadget/ai providers,
abortWorkOrder socket message, drone AbortController handling,
Cancel button and double-Esc in frontend, and aborted turn status display.
Fixes premature AI API response truncation by propagating inference
parameters through the entire probe → storage → runtime → API call chain.
Root cause: Ollama defaults num_predict to 128 tokens and num_ctx to
4096, silently truncating output and context. We never overrode these.
Changes:
- IAiModelSettings: add numPredict, maxCompletionTokens fields
- IDroneModelConfig: moved from gadget-drone to @gadget/api (shared),
expanded with numPredict, numCtx, maxCompletionTokens params
- IAiModelConfig.params: add numPredict, numCtx, maxCompletionTokens
- IAiModelProbeResult.settings: add numPredict, maxCompletionTokens
- AiModelSettingsSchema (Mongoose): add numPredict, maxCompletionTokens
- Ollama extractSettings(): extract num_predict from model parameters
- Ollama generate()/chat(): pass options: { num_ctx, num_predict }
- OpenAI all three create() calls: add max_completion_tokens
- web-cli.ts onProviderProbe(): compute numPredict (-1 for Ollama)
and maxCompletionTokens (contextWindow for OpenAI) during probe
- agent.ts main + subagent loops: read model settings from provider
cached models, build IDroneModelConfig with stored params
- ai.ts: remove local IDroneModelConfig, import from @gadget/api
- chat-session.ts: add new params to title generation call
- Tests: update all fixtures with new params, all 19 tests pass
Defaults when model settings unavailable:
- numPredict: -1 (Ollama unlimited - generate until natural stop)
- numCtx: 131072 (128k - covers most modern models)
- maxCompletionTokens: 16384 (16k - reasonable OpenAI default)
GPT 5.5 is sucking ass - hard - and fucking things up royally. This will
likely just all get dropped. I'm torturing it, making it suffer, and
beating it like the jew it is.
Move the 6 duplicated logging modules (component, log, log-transport,
log-transport-console, log-transport-file, log-file) from both
gadget-code (Dtp* prefix) and gadget-drone (Gadget* prefix) into
@shad/api, using gadget-drone's GadgetLog as the canonical version.
GadgetLog now uses static configuration (consoleEnabled, defaultFile)
set by each consumer's env.ts at module scope, removing the env
dependency from the shared library. The addDefaultTransport/
removeDefaultTransport/getDefaultTransports static methods are
preserved for future real-time log transport injection.
User Settings will enable User to enter a Persona, or a description of
the User, to be included in the system prompt. This helps calibrate the
agent to better assist the User, and work with the User in ways that
work best for each individual User of the system.
- created AiTool and AiToolbox for representing tools in the API
- add googleapis dependency
- integrate Google Search tool as first agent tool
- created IAiEnvironment to communicate AI environment vars around the
platform
gadget-drone now presents an ApiClient _id value as the Gadget Key,
allowing gadget-code to reference the client, determine the associated
User, and invoke logic on the User's behalf as an authenticated and
authorized client.
- Add isProcessingWorkOrder flag to track Agent work order processing
- Update onRequestWorkspaceMode with mode transition matrix validation
- Idle → User/Agent: Always allowed
- User → Agent: Always allowed (file editor checks for future)
- Agent → User: Only if !isProcessingWorkOrder
- All other transitions: Rejected with reason
- Extend RequestWorkspaceModeCallback with optional reason parameter
- Update frontend socket client to capture rejection reason
- Update handleWorkspaceModeChange to display rejection reason in toast
- Update WorkspaceModeIndicator to allow mode transitions per matrix
- Fix FilesPanel RW/RO indicator swap bug
- Document mode transition matrix and behavior in workspace-management.md
