Prompt file imported from Jhilam2014/orchestrator-agent-001 (
.codex/prompts/bootstrap-orchestrator.md). Copyright stays with the author.
Read AGENTS.md and ROOT_WORKSPACE_GENERATION_POLICY.md fully.
Enable orchestrator-agent mode.
Bootstrap the mandatory orchestrator infrastructure for this project.
Do not modify AGENTS.md.
Create required folders and files only now, during bootstrap.
If this project already contains orchestrator-generated folders or files from an older orchestrator-agent version, run bootstrap as an upgrade/repair pass.
Do not delete, reset, regenerate, or replace existing orchestrator artifacts.
Only create missing files and patch existing files when required.
Preserve existing agents, registry, memory, observability, graph, topology, Neo4j, D3, ChromaDB, vector sync records, capability scores, and human-review state.
Do not implement application features yet.
Create and verify the bootstrap artifacts required by AGENTS.md, including:
- Required root workspace folders.
- Local agents folder and default local execution agent if required.
- Local agent registry.
- Neo4j graph artifacts.
- Agent-to-functionality graph schema.
- Agentic System D3 page.
- OpenAI Vector Store integration if configured in
.env. - ChromaDB fallback if no vector DB is configured in
.env. - Prompt memory ingestion for orchestrator, agents, subagents, handoff prompts, validation prompts, and correction prompts.
- Observability logs.
- Token observability schema, token publisher runtime service, token plan, token events, and bootstrap token report.
- Verification scripts.
Follow ROOT_WORKSPACE_GENERATION_POLICY.md for all file placement.
After bootstrap, report:
- folders created
- files created
- files modified
- files skipped or preserved
- Neo4j status
- vector DB provider selected
- ChromaDB fallback status
- D3 Agentic System page path
- token observability schema path
- token publisher path
- token observability files created
- validation checks passed
- pending credentials or failed checks