Agent Academy Knowledge Bundle
Bundle OKF 0.2 · 20 conceitos · Three-Wheeled-Sloth-Studio/Agent-Academy
Open source Repository Open in the app JSON README (API)
About
# Agent Academy Knowledge Bundle
Generated OKF discovery index. Do not edit manually.
## Directories
* [Architecture](architecture/) - Browse Architecture knowledge and resources.
* [Data](data/) - Browse Data knowledge and resources.
* [Examples](examples/) - Browse Examples knowledge and resources.
* [Handoffs](handoffs/) - Browse Handoffs knowledge and resources.
* [Implementation](implementation/) - Browse Implementation knowledge and resources.
* [Integrations](integrations/) - Browse Integrations knowledge and resources.
* [Operations](operations/) - Browse Operations knowledge and resources.
* [Planning](planning/) - Browse Planning knowledge and resources.
* [Product](product/) - Browse Product knowledge and resources.
* [Schemas](schemas/) - Browse Schemas knowledge and resources.
* [Testing](testing/) - Browse Testing knowledge and resources.
* [Tools](tools/) - Browse Tools knowledge and resources.
* [UI](UI/) - Browse UI knowledge and resources.
## Concepts
* [Framewor
Details
- Kind
- OKF bundles
- Topic
- Developer tools
- Publisher
- three-wheeled-sloth-studio
- Origin
- okf_github
- Category
- dados
- Version
- 0.2
- Open pull requests
- 1
- Last push
- 2026-09-08T20:57:09Z
- Repository state
- ativo
- Language
- Python
- License
- NOASSERTION
- Added
- 2026-09-08 22:07:22
- Updated
- 2026-09-08 22:07:22
- Origin id
Three-Wheeled-Sloth-Studio/Agent-Academy:refs/index.md
README
# Agent Academy Refs Harness This repository contains a reusable `refs/` harness for project memory. It is intentionally project-neutral: copy it into another repository, then fill it with durable facts, decisions, workflows, architecture notes, validation commands, and handoff context for that project. The harness is designed for human collaborators and coding agents. It gives agents a predictable memory layout plus a bounded re-entry workflow so routine context resets do not require rereading the entire project history. Agent Academy is compatible with Open Knowledge Format (OKF) v0.2 while retaining stricter deterministic YAML where exact state and validation matter. The compatibility profile is pinned in `refs/okfProfile.yaml` to the canonical `GoogleCloudPlatform/open-knowledge-format` specification. ## Use 1. Copy the `refs/` directory into the target project root. 2. Read `refs/README.md`. 3. Fill the required bootstrap files listed in `refs/templatePolicy.yaml`. 4. Regenerate OKF discovery indexes with `python refs/tools/generate_okf_indexes.py`. 5. Run the validation commands in `refs/testing/validationCommands.yaml`. 6. For routine coding-agent continuation, begin with `python refs/tools/generate_agent_context.py --focus "<task>"` and load deeper context progressively only as needed. 7. Keep project facts current as implementation work changes. Generated re-entry packets are derived scratch context, not project state. Do not commit them as an alternate source of truth. Do not store secrets, API keys, passwords, tokens, or machine-only credentials in `refs/`. ## Framework Maintenance Framework authors should keep file paths stable because the `refs/` layout is the public interface for copied projects. See `refs/MAINTENANCE.md` before adding files, changing schemas, upgrading the OKF profile, or introducing migration steps.