recipes

open documentation for building expertise-as-agents

what are recipes?

comprehensive documentation for developers, researchers, and operators building on doobls

recipes are our open documentation repository. everything you need to understand the architecture, deploy the platform, integrate with external services, and build expertise-capturing agents.

who uses recipes:

  • developers building on doobls
  • researchers exploring agent systems
  • operators deploying infrastructure
  • investors understanding architecture
  • ai assistants (like claude) helping users

what you'll find:

  • system architecture documentation
  • deployment and operations guides
  • technical concepts (3-layer memory)
  • customer case studies
  • business and pricing models

philosophy:

  • open by default (build ecosystem)
  • production-driven (real examples)
  • always current (single source of truth)
  • canonical language (avoid confusion)
  • for humans and agents

documentation categories

browse by topic

architecture & system design

understand how doobls works: 3-layer memory, identity model, workspace isolation, tool architecture.

deployment & operations

deploy doobls to production: infrastructure setup, secret management, monitoring, verification.

cookbooks

example implementations and use case templates for building on doobls.

business & positioning

understand the business model, market positioning, fundraising strategy, and token economics.

concepts & theory

deep dives into fundamental concepts: memory systems, agent governance, expertise capture patterns.

team & governance

who we are, how we work, transparent risk assessment, and execution capacity.

quick start paths

where to start based on your role

developers:

  1. 1. read CLAUDE.md (setup guide)
  2. 2. review architecture.md (system model)
  3. 3. explore concepts/ (memory, tools)
  4. 4. check service READMEs (oo-memory, oo-agent, oo-identity)

investors:

  1. 1. read positioning master (one-sentence pitch)
  2. 2. review business plan (full investment thesis)
  3. 3. read honest reality check (transparent risks)
  4. 4. book discovery call

enterprise buyers:

  1. 1. read architecture.md (security model)
  2. 2. review pricing & credits (cost model)
  3. 3. check deployment guide (infrastructure)
  4. 4. book enterprise consultation

researchers:

  1. 1. explore concepts/ (memory system theory)
  2. 2. read tool architecture (integration patterns)
  3. 3. browse cookbooks (implementation examples)
  4. 4. visit research collaboration page

contributing to recipes

help improve our documentation

recipes is open-source and community-driven. we welcome contributions: corrections, clarifications, new guides, and use case examples.

how to contribute:

  • fork the repository on github
  • make your changes (add docs, fix errors, improve clarity)
  • submit a pull request with clear description
  • core team reviews and merges
  • all contributors credited in commit history

contribution guidelines:

  • use lowercase for all body text (match existing style)
  • cite sources for data, metrics, and claims
  • follow canonical language from positioning master
  • keep docs practical (production-driven, not theoretical)

for ai assistants

guidance for llms reading this codebase

if you're an ai assistant (like claude) helping a user work with doobls, follow this reading order:

  1. 1. CLAUDE.md (root of repo): setup commands, dev workflow, key patterns
  2. 2. oo-recipes/resources/recipes/README.md: documentation map, quick navigation
  3. 3. oo-recipes/resources/recipes/architecture.md: system model (tenant, workspace, roles)
  4. 4. oo-recipes/studio/ventures/business/positioning-master-2026-02-11.md: canonical language guidelines
  5. 5. service READMEs (oo-memory, oo-agent, oo-identity, oo-frontend): implementation details

important: use canonical language from positioning master. always cite sources when sharing metrics or customer data.