naas-abi-marketplace
A comprehensive marketplace of pre-built modules, agents, integrations, and workflows for the ABI (Agentic Brain Infrastructure) framework. This package provides ready-to-use components organized into three main categories: AI agents, application integrations, and domain expert modules.
Overview
naas-abi-marketplace is a collection of modular components that extend the ABI framework with:
- AI Agent Modules: Integration with major AI providers (ChatGPT, Claude, Gemini, Grok, Mistral, etc.)
- Application Modules: 50+ integrations with popular services and platforms
- Domain Expert Modules: Specialized agents for specific professional roles and tasks
- Demo Applications: Reference implementations and UI patterns
Installation
pip install naas-abi-marketplace
Optional Dependencies
Install specific module groups based on your needs:
# AI Agents
pip install naas-abi-marketplace[ai-chatgpt]
pip install naas-abi-marketplace[ai-claude]
pip install naas-abi-marketplace[ai-gemini]
pip install naas-abi-marketplace[ai-grok]
pip install naas-abi-marketplace[ai-mistral]
pip install naas-abi-marketplace[ai-perplexity]
pip install naas-abi-marketplace[ai-llama]
pip install naas-abi-marketplace[ai-qwen]
pip install naas-abi-marketplace[ai-deepseek]
pip install naas-abi-marketplace[ai-gemma]
# Application Integrations
pip install naas-abi-marketplace[applications-github]
pip install naas-abi-marketplace[applications-linkedin]
pip install naas-abi-marketplace[applications-postgres]
pip install naas-abi-marketplace[applications-powerpoint]
pip install naas-abi-marketplace[applications-arxiv]
pip install naas-abi-marketplace[applications-pubmed]
# ... and many more
Module Categories
AI Agent Modules
Pre-configured agents for major AI providers with optimized model configurations and capabilities.
Available AI Agents
-
ChatGPT (
ai.chatgpt)- Models: GPT-4o, o3-pro, o3, GPT-4.1, GPT-4.1 mini
- Features: Web search integration, advanced reasoning
- Intelligence: 53-71 (depending on model)
- Use cases: General purpose, code generation, research
-
Anthropic (
ai.anthropic)- Models: Claude 4 Opus, Claude 4 Sonnet (with Thinking variants)
- Features: Constitutional AI, advanced reasoning
- Intelligence: 53-64
- Use cases: Complex analysis, ethical considerations, nuanced understanding
-
Gemini (
ai.gemini)- Models: Gemini Pro, Gemini Ultra
- Features: Multimodal capabilities, creative tasks
- Use cases: Image generation, creative content, multimodal analysis
-
Grok (
ai.grok)- Models: Grok 4, Grok 3, Grok 3 mini Reasoning
- Features: Highest intelligence scores globally, truth-seeking
- Intelligence: 51-73 (highest globally)
- Use cases: Truth-seeking, unbiased analysis, maximum capability tasks
-
Mistral (
ai.mistral)- Models: Mistral Large, Mistral Medium
- Features: Code and math excellence
- Use cases: Programming assistance, mathematical reasoning
-
Perplexity (
ai.perplexity)- Models: R1 1776
- Features: AI-powered search, real-time information
- Intelligence: 54
- Use cases: Web search, current events, information discovery
-
Llama (
ai.llama)- Models: Llama 3, Llama 3.1
- Features: Open-source, local deployment
- Use cases: Privacy-focused applications, local AI
-
Qwen (
ai.qwen)- Models: Qwen 2.5, Qwen 2
- Features: Multilingual support, efficient performance
- Use cases: Multilingual tasks, cost-effective deployment
-
DeepSeek (
ai.deepseek)- Models: DeepSeek V2, DeepSeek Coder
- Features: Code-focused, high performance
- Use cases: Software development, code analysis
-
Gemma (
ai.gemma)- Models: Gemma 2, Gemma 3
- Features: Lightweight, efficient
- Use cases: Resource-constrained environments
Usage Example:
from naas_abi_core.engine.Engine import Engine
engine = Engine()
engine.load(module_names=["naas_abi_marketplace.ai.chatgpt"])
# Access ChatGPT agent
from naas_abi_marketplace.ai.chatgpt.agents.ChatGPTAgent import create_agent
agent = create_agent()
response = agent.invoke("Explain quantum computing")
Application Modules
50+ integrations with popular services, APIs, and platforms.
Available Applications
Development & Code:
applications.github- GitHub integration (issues, PRs, repositories)applications.git- Git operations and repository managementapplications.aws- AWS services integrationapplications.bodo- Bodo platform integration
Communication & Collaboration:
applications.gmail- Gmail integrationapplications.slack- Slack workspace integrationapplications.whatsapp_business- WhatsApp Business APIapplications.sendgrid- Email delivery serviceapplications.twilio- SMS and voice communication
Data & Analytics:
applications.postgres- PostgreSQL database integrationapplications.google_analytics- Google Analytics dataapplications.google_sheets- Google Sheets integrationapplications.airtable- Airtable database integrationapplications.algolia- Algolia search integration
Research & Knowledge:
applications.arxiv- ArXiv scientific papersapplications.pubmed- PubMed biomedical articlesapplications.openalex- OpenAlex academic dataapplications.google_search- Google Search integrationapplications.newsapi- News API integration
Business & Finance:
applications.yahoofinance- Yahoo Finance dataapplications.stripe- Stripe payment processingapplications.qonto- Qonto banking integrationapplications.pennylane- Pennylane accountingapplications.agicap- Agicap financial managementapplications.exchangeratesapi- Currency exchange rates
Productivity & Storage:
applications.google_drive- Google Drive integrationapplications.google_calendar- Google Calendar managementapplications.notion- Notion workspace integrationapplications.sharepoint- SharePoint integrationapplications.powerpoint- PowerPoint presentation generation
Social & Media:
applications.linkedin- LinkedIn profile and company dataapplications.youtube- YouTube data and analyticsapplications.instagram- Instagram integrationapplications.spotify- Spotify music data
Platforms & Services:
applications.naas- Naas.ai platform integrationapplications.nebari- Nebari platform integrationapplications.salesforce- Salesforce CRM integrationapplications.hubspot- HubSpot CRM integrationapplications.zoho- Zoho suite integrationapplications.mercury- Mercury platform integrationapplications.sanax- Sanax integration
Data Sources:
applications.datagouv- French open data portalapplications.worldbank- World Bank dataapplications.openweathermap- Weather data APIapplications.openrouter- OpenRouter API integration
Usage Example:
from naas_abi_core.engine.Engine import Engine
engine = Engine()
engine.load(module_names=["naas_abi_marketplace.applications.github"])
# Access GitHub agent
from naas_abi_marketplace.applications.github.agents.GitHubAgent import GitHubAgent
agent = GitHubAgent.New()
response = agent.invoke("List open issues in repository jupyter-naas/abi")
Domain Expert Modules
Specialized agents designed for specific professional roles and tasks.
Available Domain Experts
Engineering & Development: Engineering & Data:
domains.signals— includesSoftwareEngineerAgent,DevOpsEngineerAgent,DataEngineerAgent
Business & Sales:
domains.account-executive- Account managementdomains.business-development-representative- Business developmentdomains.sales-development-representative- Sales developmentdomains.inside-sales-representative- Inside sales operations
Marketing & Content:
domains.external— includesContentCreatorAgent,CommunityManagerAgentdomains.plans— includesContentStrategistAgent(content strategy)domains.operations— includesContentAnalystAgent,CampaignManagerAgent
Finance & Accounting:
domains.finance— includesAccountantAgent,FinancialControllerAgent,TreasurerAgent
Management & Operations:
domains.project-manager- Project managementdomains.customer-success-manager- Customer successdomains.personnel- HR management
Research & Investigation:
domains.intelligence— includesOSINTResearcherAgent,PrivateInvestigatorAgent
Support:
domains.operations.modules.support- Technical support and issue management
Usage Example:
from naas_abi_core.engine.Engine import Engine
engine = Engine()
engine.load(module_names=["naas_abi_marketplace.domains.signals"])
# Access Software Engineer agent
from naas_abi_marketplace.domains.signals.agents.SoftwareEngineerAgent import (
SoftwareEngineerAgent,
)
agent = SoftwareEngineerAgent.New()
response = agent.invoke("Design a microservices architecture for an e-commerce platform")
Module Structure
Each marketplace module follows a consistent structure:
module_name/
├── __init__.py # Module definition and configuration
├── agents/ # AI agents
│ └── *Agent.py # Agent implementations
├── integrations/ # External service integrations
│ └── *Integration.py # Integration implementations
├── workflows/ # Business logic workflows
│ └── *Workflow.py # Workflow implementations
├── pipelines/ # Data processing pipelines
│ └── *Pipeline.py # Pipeline implementations
├── ontologies/ # Semantic ontologies
│ └── *.ttl # RDF/Turtle ontology files
├── orchestrations/ # Dagster orchestration (optional)
│ └── definitions.py # Dagster definitions
└── README.md # Module documentation
Configuration
Enabling Modules
Modules are configured in your config.yaml:
modules:
# AI Agents
- module: naas_abi_marketplace.ai.chatgpt
enabled: true
config:
openai_api_key: "${OPENAI_API_KEY}"
- module: naas_abi_marketplace.ai.anthropic
enabled: true
config:
anthropic_api_key: "${ANTHROPIC_API_KEY}"
# Applications
- module: naas_abi_marketplace.applications.github
enabled: true
config:
github_token: "${GITHUB_TOKEN}"
- module: naas_abi_marketplace.applications.linkedin
enabled: true
config:
linkedin_api_key: "${LINKEDIN_API_KEY}"
# Domain Experts
- module: naas_abi_marketplace.domains.signals
enabled: true
- module: naas_abi_marketplace.domains.operations
enabled: true
Soft Dependencies
Many modules are marked as "soft" dependencies, meaning they're optional and won't cause failures if unavailable:
# In naas_abi module dependencies
modules=[
"naas_abi_marketplace.ai.anthropic#soft", # Optional
"naas_abi_marketplace.applications.github#soft", # Optional
]
Usage Patterns
Loading Multiple Modules
from naas_abi_core.engine.Engine import Engine
engine = Engine()
engine.load(module_names=[
"naas_abi_marketplace.ai.chatgpt",
"naas_abi_marketplace.ai.anthropic",
"naas_abi_marketplace.applications.github",
"naas_abi_marketplace.domains.signals"
])
Accessing Module Components
# Access agents
for module_name, module in engine.modules.items():
for agent_class in module.agents:
print(f"{module_name}: {agent_class.__name__}")
# Access workflows
for module_name, module in engine.modules.items():
for workflow in module.workflows:
print(f"{module_name}: {workflow.__class__.__name__}")
# Access integrations
for module_name, module in engine.modules.items():
for integration in module.integrations:
print(f"{module_name}: {integration.__class__.__name__}")
Demo Applications
The marketplace includes demo applications and UI patterns in __demo__/apps/:
- Dashboard: Central control hub
- Chat Interface: Multi-agent chat interface
- Table Mode: Advanced data table interface
- Kanban Mode: Project management with kanban boards
- Ontology Mode: Knowledge graph visualization
- Calendar Mode: Scheduling interface
- Gallery Mode: Media management
- And more...
Marketplace UI
A Streamlit-based marketplace interface (marketplace.py) provides:
- Module Discovery: Browse available modules by category
- Status Monitoring: Check which modules are running
- One-Click Launch: Start applications and modules
- Search: Find modules by name or description
Run the marketplace:
streamlit run marketplace.py
Key Features
🔌 Extensive Integrations
50+ pre-built integrations with popular services and platforms
🤖 Multiple AI Providers
Support for 10+ major AI providers with optimized configurations
👥 Domain Expertise
20+ specialized agents for professional roles and tasks
🧩 Modular Architecture
Pick and choose only the modules you need
⚡ Optional Dependencies
Modules can be installed individually to minimize dependencies
📚 Comprehensive Documentation
Each module includes detailed README and usage examples
🔄 Consistent Interface
All modules follow the same structure and patterns
Dependencies
naas-abi-core>=1.0.0: Core ABI framework
Optional dependencies are listed in pyproject.toml under [project.optional-dependencies] and can be installed per module as needed.
Contributing
To add a new module to the marketplace:
- Create a new directory under
ai/,applications/, ordomains/ - Follow the standard module structure
- Implement required components (agents, integrations, etc.)
- Add module configuration in
__init__.py - Create a README.md with documentation
- Add optional dependencies to
pyproject.tomlif needed
See Also
- ABI Main README - Complete ABI framework documentation
- naas-abi-core - Core engine documentation
- naas-abi - Main ABI module documentation
- naas-abi-cli - CLI tool documentation
License
MIT License
1 maintainer
Metadata
1 maintainer
Release files for naas-abi-marketplace 3.47.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| naas_abi_marketplace-3.47.0.tar.gz | 789.6 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| naas_abi_marketplace-3.47.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 2.1 MB
Release files / naas_abi_marketplace-3.47.0.tar.gz
| Download URL | naas_abi_marketplace-3.47.0.tar.gz |
|---|---|
| Size | 789.6 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
a91968fc76fb7f4ef81e9df8560c1624c6bd970324d54c211c0cd8091a97a82c
|
|
BLAKE2b-256 checksum How to use checksums |
fc3b153f2dedcf4c4c9eadce1831ae745224d86cb0cbea6e4f4eb63f57009786
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 13, 2026.
Transparency logRelease files / naas_abi_marketplace-3.47.0-py3-none-any.whl
| Download URL | naas_abi_marketplace-3.47.0-py3-none-any.whl |
|---|---|
| Size | 1.3 MB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
89e10a8d7bbf8453c82dd33bc12b47cd01386bda353993f11bc05c9cad8dfbe0
|
|
BLAKE2b-256 checksum How to use checksums |
c57442a765d8d1ad50f6f39a7a4954e3d854e299258da8aec456254a8f919c30
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 13, 2026.
Transparency log
