MardesOS
Resources

Data Source Registry

A safe-usage registry for external and client-owned sources that may support future analysis. No scraping or connector behavior is active on the commercial website.

Governance resource

Honest staging content for internal evaluation. No fake proof, live agents, scraping, billing, CRM, or backend automation is introduced.

Source boundaries

Research sources require permission, license review, and human approval.

The registry explains what each source may be useful for, what should not be done, and whether a future connector is planned.

Kaggle

Public datasets for exploratory learning, market examples, and model evaluation planning.

Planned source

Safe usage approach

Use only datasets with reviewed licenses and documented provenance.

What should not be done

Do not copy restricted datasets, imply ownership, or use data without checking license terms.

License / permission note

License-sensitive; every dataset requires manual review.

Future connector status

Future connector not active.

Hugging Face

Model and dataset discovery, benchmark research, and AI workflow planning.

Manual review required

Safe usage approach

Review dataset cards, model licenses, intended use, and safety limitations before use.

What should not be done

Do not claim active model integration or use restricted models without approval.

License / permission note

License and acceptable-use review required.

Future connector status

Future research source, not an active connector.

Meta Ad Library

Public ad transparency research and competitive messaging review where permitted.

API-first

Safe usage approach

Prefer approved APIs and manual review; cite source context carefully.

What should not be done

Do not scrape in violation of terms or copy creative as owned material.

License / permission note

Terms-of-use and API policy review required.

Future connector status

Planned research path only.

Swiped.co

Manual creative inspiration and copywriting pattern research.

License-sensitive

Safe usage approach

Use as inspiration with attribution discipline; do not copy protected examples.

What should not be done

Do not scrape, republish, or present examples as original work.

License / permission note

Copyright-sensitive; manual review required.

Future connector status

No connector planned without approval.

Public websites

Public positioning research, content structure review, and market education.

Manual review required

Safe usage approach

Use public pages for manual research and summarize findings without copying content.

What should not be done

Do not scrape at scale, bypass restrictions, or copy copyrighted page content.

License / permission note

Robots, terms, copyright, and fair-use considerations apply.

Future connector status

No active scraping.

Client-owned data

Approved campaign analysis, CRM review, reporting, and workflow design.

Client-owned only

Safe usage approach

Use only with explicit client permission, access controls, and agreed purpose.

What should not be done

Do not mix client data into reusable templates or public materials.

License / permission note

Client authorization and data-processing terms required.

Future connector status

Future controlled integration path.

Internal CRM data

Lifecycle operations, account context, handoff planning, and reporting.

Future connector status

Safe usage approach

Use only approved fields with consent, retention, and access controls.

What should not be done

Do not enrich, export, or automate outreach without governance approval.

License / permission note

Internal data governance and role-based access required.

Future connector status

Not active on the commercial website.

Data use remains approval-first.

Future connectors should be API-first where possible, documented, permission-aware, and reviewed before use.

View case library