Use Cases: Start with the Outcome

Founders and small teams: Keep track of every decision your team makes, and why shows how people and agents recover the evidence, approval, consequences and open questions behind a decision and continue from the current answer.

Teams that send documents to be signed: Write it with your agent, get it signed where it lives shows how an agent's draft is reviewed, sent for signature from the web app and kept with its signed PDF.

Teams onboarding new people: Answer new people's questions from your onboarding material shows how a team keeps its onboarding material in one organization, and how a new person's own agent, or the team's own assistant or MCP server with a read-only key, answers from it.

Firms sharing work with another firm: Share work confidentially between your firm and another shows how a creator, agency, founder or consultant prepares only what an outside partner needs in a separate private organization, and invites them to read and take part without connecting an agent.

People running bounded deep research: A Throwaway Organization for Deep Research shows how an investigation can expand freely for hours or days, preserve sources and intermediate knowledge, then distill only the valuable one, two, or three documents into a durable personal, team, or public organization.

Designers and product teams working with agents: Design new screens with your agent on your own design system, and review them on a Lens shows how an agent builds clickable HTML mockups from your real design system, lays the alternatives out on a Lens beside the brief, and keeps each decision in documents for the next screen and for the code.

Software companies keeping their docs current: Keep your public docs current from what your team knows shows how a private organization holds what the team knows, a public one holds what outsiders read, and a Loop turns each delivered change into an approved update that outside agents and your docs site both read.

Builders of support and voice agents: Give your support or voice agent a knowledge base per customer shows how a product keeps each customer's knowledge in its own organization and lets its agent search it when a question comes in.

Teams coordinating agent implementation: A Shared Implementation Journal for People and Agents shows how one living coordination document makes an agent's plan, active work, findings, approval gates, canonical updates, and verification visible to collaborators and later agents.

Engineering teams reviewing with agents: Review every change with several agents, from anywhere shows how reviews become documents any agent writes for an exact commit, several different agents review the same change, and a CI step turns their verdicts into a check.

Teams testing their product with agents: Test your product with your own agents shows how your own agents use the product like people do and keep each run's findings where your coding agent fixes and rechecks them.

Teams running many agents in parallel: A Shared Claim Board for People and Agents shows how one document with a "Taken by" column lets people and agents claim work before starting, record progress and results, and add new work as rows, so nobody starts the same item twice.

People designing agent systems: Loop and Graph Engineering shows how Oddlynew can serve as the durable control plane while chosen agents and runtimes own reasoning, triggers, and execution.

Teams operating or investigating agent fleets: Understand and Steer Agent Fleets shows how several analysis agents can turn authorized telemetry, transcripts, external receipts, competing hypotheses, and coverage gaps into one durable operating model that people and later agents can inspect, challenge, and steer.

Content creators: A Durable Editorial System for Content Creators shows how ideas, drafts, audience signals, editorial decisions, and published learning can become one continuing system.

Faceless and automated channel operators: Faceless Channel Operations shows how YouTube evidence, research, production tools, voice, video generation, maintained expert knowledge, and channel-specific learning can become one operating loop rather than a stack of disconnected generation tools.

Marketers and growth operators: A Campaign Learning Loop for Marketers shows how campaign evidence becomes reusable decisions and better next experiments.

Founders, sales and growth teams doing outbound: Find the right accounts, research them, reach out, and learn from every campaign shows how agents find and research accounts from your own data sources, score them against your ideal customer, draft outreach for a person to send, and write what each campaign taught back into who to target and what to say.

People doing specialist work without the credential, and the companies that hire them: Do specialist work with an expert organization behind your agent shows how a capable generalist and their agent work from an organization that holds a field's rules and procedures with their sources, keep client cases apart, and leave the knowledge better after each case. Anyone who builds such an organization can also publish it for others.

Medical research and teaching groups: Coordinating Medical Research and Teaching Groups shows how specialist research subgroups can contribute evidence to shared papers while anatomy tutoring groups produce reviewed presentations and educational material under explicit access boundaries.

several people or agents need the same context;

decisions must remain explainable after the conversation ends;

evidence and conclusions change over time;

a recurring workflow should improve from its prior runs;

the audience or membership boundary matters;

the volume of knowledge is too large for one prompt.