What changed
Setup keeps your keys
the README, QUICKSTART and UPGRADING setup steps no longer overwrite an existing .env when you run them again.
Package version fixed
pyproject.toml reports the release version; in 2.0.0 it still said 1.1.1.
Commercial license ships
LICENSE is in the repository and the README license line points to it and to the Tegra terms.
Nothing else changes
78 working skills, 6 slash commands, 34 specialist agents and 17 enforcement hooks, exactly as in 2.0.0.
Release notes
This is a small maintenance release on top of 2.0.0. It changes no workflow, skill, agent or hook, and it needs no re-clone: run /update or bash scripts/update.sh in your copy.
Setup is safe to repeat
The setup steps in README.md, QUICKSTART.md and UPGRADING.md now copy the example environment file with cp -n .env.example .env. Running them again keeps the .env you already filled in instead of replacing it with the blank example.
Package metadata matches the release
pyproject.toml now carries the release version, so pip show meta-ads-ai-agentic-system and any tooling that reads the package version report 2.0.1. Version 2.0.0 shipped with the older 1.1.1 value there.
License
The repository includes LICENSE, the commercial license for purchasers of the full Meta Ads AI Agentic System: internal team use by the purchasing business, including client accounts it is authorized to manage, with no redistribution, resale or sublicensing. It mirrors the system team license in the Tegra terms.
Unchanged
- 78 working skills plus 28 short notices for retired names, 6 slash commands, 34 specialist agents and 17 enforcement hooks.
- Still not in this edition:
/atlas, the performance audits and the insights service they read, and the/fb-minerunner.
Before adopting this release
Open your purchased repository’s compatibility and update guidance. Preserve local configuration and repeat one representative task with a reviewer. Public notes do not establish your entitlement, installed version or account permissions.