No description
- Nickel 99.1%
- Just 0.9%
| Filename | Latest commit message | Latest commit date |
|---|---|---|
adr-018 (Accepted): LLM providers are reached through typedialog-llm, with tool calling and provider-reported usage behind LLMClient; tool-calling turns run through LLMRouter::chat_with_budget. Constraints: provider access only through typedialog-llm, tool turns only through the router, no credentials in URLs. Engages cost-vs-capability and agent-autonomy-vs-budget-control, held by budgeting every turn. adr-006 (Superseded by adr-018): rig-core was declared for tool calling and never imported. |
||
| .ontoref | ||
| assets | ||
| .gitignore | ||
| CHANGELOG.md | ||
| justfile | ||
| README.md | ||
Development Flows When Teams and AI Orchestrate
Specialized agents orchestrate pipelines for design, implementation, testing, documentation and deployment. Agents learn from history and optimize costs automatically. 100% self-hosted.
vapora/
├── .ontoref/ protocol spine — ontology, ADRs, reflection, positioning
├── assets/ canonical brand assets — logos, SVG marks
├── code/ git repo (18-crate Rust workspace, Leptos UI, migrations)
├── outreach/ web content, presentations
└── vault/ private strategy, SOPS-encrypted
Quick start
cd code
cargo build --workspace
just ci-quick
Start the full stack:
# Prerequisites: SurrealDB + NATS
docker compose up -d
# Backend: http://localhost:8001
# Frontend: http://localhost:3000
Sub-repos
| Directory | Remote | Purpose |
|---|---|---|
code/ |
https://repo.jesusperez.pro/jesus/Vapora | Implementation — Rust workspace, UI, infra |
outreach/ |
TBD | Web content, presentations |
vault/ |
TBD | Private strategy, SOPS-encrypted |
Full technical documentation → code/README.md