We think with AI and create new experiences through creativity. From personal reflection with OpenAlarm to portfolio-based research with OpenStockAI, explore how our ideas become products.
01 / Available now
OpenAlarm
Your patterns. Your feelings. A reply that connects.
OpenAlarm considers recurring patterns and emotional context in your diary to shape a personal AI reply. Explore suggestions based on alarm use, revisit past entries and trace place-linked plans on a footprint map.
Records are handled within the user's account scope. Diary input is separated from instructions, and replies containing secrets or internal information are rejected before storage with a security alert.
OpenStockAI puts a question in portfolio context, routes it through the appropriate research path, and organizes the response around evidence, dates and risk. It is research support, not a return promise.
Research agents work alongside a security team. Checks identify exfiltration and privilege-bypass signals, connect request blocking to redacted audit records, and support review.
AI that divides the work.
Boundaries that stay in place.
Orchestrators connect tools and agents, while product-specific safeguards check access, inputs and outputs. Explore where detection, blocking and limits on sensitive-data transmission take place.
Research agents work alongside a security team. Checks identify exfiltration and privilege-bypass signals, connect request blocking to redacted audit records, and support review.
Architecture walkthrough · not live security telemetry
OpenStockAI / 01
Input
Question · portfolio context
Selected work
01Research tools
02Specialist agents
03Synthesis & review
Security checks at request time
AI input protection and training policy
Google API user data is blocked from designated external-model inputs. Supported paths disable response storage, while security analysis and records redact sensitive information.
Read the protection scope
The security team combines server rules and optional AI-assisted review. Detected signals and operating settings determine logging, request blocking and temporary blocks.
A harness applies safeguards around AI execution. Transmission controls, redaction and output checks reduce exposure of sensitive information; provider retention and training policies are managed separately.
Inside the work / 03
Behind every result, a process.
Choose a product and follow its process. See what each stage receives, what it does and what it produces.
OpenStockAI / Deep team research
One question. Multiple perspectives. Evidence reviewed again.
Follow the coded team research path: seven specialist roles, department reports, decision analysis, conditional debate and audit.
Workflow walkthrough · On narrow screens, scroll horizontally to explore.
Workflow walkthrough / 01 — 14
OpenStockAI
Question & holdings
Input
Question · holdings
Work
Read the question in portfolio context.
Output
Results and evidence for the next stage
OpenStockAI
Intent & routing
Input
Question intent
Work
Choose the appropriate response or team research path.
Output
Results and evidence for the next stage
OpenStockAI
Research agenda
Input
Question · selected path
Work
Set the subject and agenda for the team.
Output
Results and evidence for the next stage
OpenStockAI
Deep financials
Input
Symbols · agenda
Work
Collect financial evidence for the research.
Output
Results and evidence for the next stage
OpenStockAI
Lighter response
Input
Lighter questions
Work
Use a separate response path. Playback below illustrates deeper team research.
Output
Results and evidence for the next stage
OpenStockAI
Company analysis
Input
Agenda · evidence
Work
Use the specialist's tools and analysis to prepare an opinion. Concurrency is bounded.
Output
Results and evidence for the next stage
OpenStockAI
Market research
Input
Agenda · evidence
Work
Use the specialist's tools and analysis to prepare an opinion. Concurrency is bounded.
Output
Results and evidence for the next stage
OpenStockAI
Competitors
Input
Agenda · evidence
Work
Use the specialist's tools and analysis to prepare an opinion. Concurrency is bounded.
Output
Results and evidence for the next stage
OpenStockAI
Valuation
Input
Agenda · evidence
Work
Use the specialist's tools and analysis to prepare an opinion. Concurrency is bounded.
Output
Results and evidence for the next stage
OpenStockAI
Future value
Input
Agenda · evidence
Work
Use the specialist's tools and analysis to prepare an opinion. Concurrency is bounded.
Output
Results and evidence for the next stage
OpenStockAI
Risk
Input
Agenda · evidence
Work
Use the specialist's tools and analysis to prepare an opinion. Concurrency is bounded.
Output
Results and evidence for the next stage
OpenStockAI
Performance
Input
Agenda · evidence
Work
Use the specialist's tools and analysis to prepare an opinion. Concurrency is bounded.
Output
Results and evidence for the next stage
OpenStockAI
Department reports
Input
Specialist opinions
Work
Synthesize opinions by department, handling missing evidence and failures.
Output
Results and evidence for the next stage
OpenStockAI
Hybrid research
Input
Department reports · agenda
Work
Enrich context through the conditional hybrid research stage.
Output
Results and evidence for the next stage
OpenStockAI
Decision analysis
Input
Research and reports
Work
Run strategy, trading, fund and portfolio analysis, then synthesize department reports.
Output
Results and evidence for the next stage
OpenStockAI
Bull case
Input
Research evidence
Work
For eligible single-symbol research, examine the bull case.
Output
Results and evidence for the next stage
OpenStockAI
Bear case
Input
Research evidence
Work
Examine the bear case in parallel with the bull case.
Output
Results and evidence for the next stage
OpenStockAI
Debate judgment
Input
Bull · bear cases
Work
Compare arguments and record unresolved conflicts. Fast, comparison and portfolio reviews skip debate.
Output
Results and evidence for the next stage
OpenStockAI
CIO synthesis
Input
Reports · discussion
Work
Combine research and discussion into the final synthesis.
Output
Results and evidence for the next stage
OpenStockAI
Approval gate
Input
Synthesis · settings
Work
When enabled and sensitive proposals exist, await administrator approval. This does not mean trade execution.
Output
Results and evidence for the next stage
OpenStockAI
Quality & risk audit
Input
Synthesis · evidence
Work
Audit quality and risk. Fast and deep modes use different review methods.
Output
Results and evidence for the next stage
OpenStockAI
Revise synthesis
Input
Audit feedback
Work
Revise the synthesis when needed and run another audit.
Output
Results and evidence for the next stage
OpenStockAI
Report & evidence
Input
Reviewed result
Work
Format the report with evidence, dates, risks and limitations.
Output
Results and evidence for the next stage
Connected stages
A workflow explanation, not a live run. Playback illustrates deeper research eligible for debate; not every question follows this path. Concurrency, enrichment, debate, approval and revision depend on settings and evidence. Errors can trigger retries or limited results.
In the works
OpenDataLabs
Connect verified documents and handwriting to shipping, delivery and payment schedules. Track deadlines and completion with source evidence and human review. In development.
A Privacy Harness checks task-specific allowlists before information reaches an external model. It limits unnecessary source and identifying data and blocks policy-violating requests.
