Repository navigation
ADR 0002 Core and CLI only
pavelravvich edited this page Sep 24, 2026
·
1 revision
- Status: Accepted
- Issue: #12 (triage of triggers: #5–#11, closed as not planned)
Triage for running agents from outside the CLI produced a service: HTTP server, async runs with a run store and callbacks, chat memory, in-process scheduler, concurrency limits, SSE, Docker. Each of these depends on how a concrete project is deployed and called; in a template they are weight that every derived project inherits and has to remove or adapt.
The template contains the Agent Core (config, graph, routers, agents, FinOps, typing, tests)
and one entry point — the CLI (npm start -- "task"). The public API for any other entry
point is runAgent(input, deps) with createAppDeps().
Webhooks, chat, schedules, run storage and deployment are built in the project created from
the template, as thin adapters over runAgent.
- The template stays small and has no infrastructure opinions.
- A scheduled run needs no code: any external cron can call the CLI.
- Every project writes its own adapter; the daily cap, run budget and cost report come with
runAgentfor free, so adapters stay thin.
AGBoss
- Home
- Delivery Pipeline
- Architecture
- Configuration
- Environments
- Workflow — the paths
- Path scope
- Workflow as a step
- Parallel — quorums and arbiters
- Routing patterns
- Rollback
- Routers
- Guards
- CLI — agboss
- App — @Application and createApp
- Several instances
- Health checks
- Context and Authorization
- Testing
- The test split
- Suites and Comparisons
- Components
- Naming
- Standard DTOs
- Examples: job-scout, flow-patterns
- Agents — the loop
- Model providers
- Tools
- Knowledge bases (RAG)
- FinOps
- Tracing
- Observability
- Terns and evaluation
- Profiles and comparisons
- Pauses — pending, resume by id
- Pause seam
- Decisions (ADR)