Name: task-large
Author: Jhilam2014

Prompt file imported from Jhilam2014/orchestrator-agent-001 (.codex/prompts/task-large.md). Copyright stays with the author.

Read AGENTS.md and ROOT_WORKSPACE_GENERATION_POLICY.md fully.

Enable orchestrator-agent mode.

Task:

Task size: large.

Rules:

Do not modify AGENTS.md.
Preserve existing features.
Perform project discovery before implementation.
Build execution topology.
Create or reuse specialist agents.
Break the work into tasks and assign to agents.
Update local agent files under agents/generated/.
Update Neo4j graph database artifacts for agents, workflows, components, APIs, services, files, and functionalities.
Update Agentic System D3 page.
Store redacted prompts, agent decisions, execution summaries, and validation summaries in vector memory.
Use ChromaDB fallback if OpenAI Vector Store is missing.
Follow ROOT_WORKSPACE_GENERATION_POLICY.md.

Validation:

Run lint/build/test if available.
Validate critical workflows.
Check regression risk.
If a command cannot run, explain why.

After completion, report:

workflow created
agents used or created
tasks completed
files modified
validation result
Neo4j update status
vector memory provider used
Agentic System D3 page path

Read AGENTS.md and ROOT_WORKSPACE_GENERATION_POLICY.md fully. Enable orchestrator-agent mode. Task: <write your task here> Task size: large. Rules: - Do not modify AGENTS.md. - Preserve existing features. - Perform project discovery before implementation. - Build execution topology. - Create or reuse specialist agents. - Break the work into tasks and assign to agents. - Update local agent files under agents/generated/. - Update Neo4j graph database artifacts for agents, workflows, components, APIs, services, files, and functionalities. - Update Agentic System D3 page. - Store redacted prompts, agent decisions, execution summaries, and validation summaries in vector memory. - Use ChromaDB fallback if OpenAI Vector Store is missing. - Follow ROOT_WORKSPACE_GENERATION_POLICY.md. Validation: - Run lint/build/test if available. - Validate critical workflows. - Check regression risk. - If a command cannot run, explain why. After completion, report: - workflow created - agents used or created - tasks completed - files modified - validation result - Neo4j update status - vector memory provider used - Agentic System D3 page path

Install task-large and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

{ "prompt": "Read AGENTS.md and ROOT_WORKSPACE_GENERATION_POLICY.md fully.\n\nEnable orchestrator-agent mode.\n\nTask: <write your task here>\n\nTask size: large.\n\nRules:\n- Do not modify AGENTS.md.\n- Preserve existing features.\n- Perform project discovery before implementation.\n- Build execution topology.\n- Create or reuse specialist agents.\n- Break the work into tasks and assign to agents.\n- Update local agent files under agents/generated/.\n- Update Neo4j graph database artifacts for agents, workflows, components, APIs, services, files, and functionalities.\n- Update Agentic System D3 page.\n- Store redacted prompts, agent decisions, execution summaries, and validation summaries in vector memory.\n- Use ChromaDB fallback if OpenAI Vector Store is missing.\n- Follow ROOT_WORKSPACE_GENERATION_POLICY.md.\n\nValidation:\n- Run lint/build/test if available.\n- Validate critical workflows.\n- Check regression risk.\n- If a command cannot run, explain why.\n\nAfter completion, report:\n- workflow created\n- agents used or created\n- tasks completed\n- files modified\n- validation result\n- Neo4j update status\n- vector memory provider used\n- Agentic System D3 page path", "variables": [], "notes": "Read AGENTS.md and ROOT_WORKSPACE_GENERATION_POLICY.md fully.", "metadata": { "source": { "provider": "github-codex-prompts", "repository": "https://github.com/Jhilam2014/orchestrator-agent-001", "path": ".codex/prompts/task-large.md", "ref": "4e53fee8e701cb31679fb0ac03220bfc41c8d022", "url": "https://github.com/Jhilam2014/orchestrator-agent-001/blob/4e53fee8e701cb31679fb0ac03220bfc41c8d022/.codex/prompts/task-large.md", "key": "Jhilam2014/orchestrator-agent-001/.codex/prompts/task-large.md" } } }

task-large

About

Use it

Install task-large and it becomes one more candidate the router can pick - when it fits.

Manifest

Reviews