agentleFS
SafetyMPSafetyMP/HR-ERP/.cursor/rules/orchestrator-hr-erp.mdc

HR-ERP / rules

Cursor rule1 starsChanged 6 months agoApache-2.0View on GitHub

HR ERP orchestrator — hooks + native runtime (Cursor 3)

The file

Files
orchestrator-hr-erp.mdc
---
description: HR ERP orchestrator — hooks + native runtime (Cursor 3)
alwaysApply: true
---

# HR ERP orchestrator

**Team map:** [agent-team-map.md](../../docs/meta/agent-team-map.md) · **Process:** [hooks.json](../hooks.json) · **Invariants:** [repo-boundaries.mdc](repo-boundaries.mdc) · **Skills:** `@skill-router` · `~/.cursor/skills/README.md`

Global core: `~/.cursor/rules/core-orchestrator.mdc` + `core-dynamic-skills.mdc` (advisory when `hookEnforcement: true`).

## Quick start

0. Collaboration phases 1–3: `@hr-human-collaboration` + [collaboration-plan.md](../../specs/templates/collaboration-plan.md)
1. Load [agent-team-map.md](../../docs/meta/agent-team-map.md) (team roster)
2. `npm run governance:lint` → tier + required lanes
3. `npm run governance:plan` → DAG JSON for handoff / hooks
4. `/multitask` · `/worktree` (DDL) · `/best-of-n` (T4 repair) per `@hr-orchestration-lanes`
5. T2+: `specs/**/orchestrator-handoff.json` with `delegatedTaskPlan` matching Required lanes
6. `npm run governance:ci` before merge (PRs)

**Hook rollout:** counsel-before-builder denies T3+ builder Tasks from `preToolUseDenyT3From` ([hook-mode.json](../governance/hook-mode.json)). Check: `npm run governance:hooks:status`.

Manifest v4: [governance-manifest.yaml](../governance/governance-manifest.yaml) · ADRs [0016–0020](../../specs/alignment/decisions/0016-agent-harness-foundation.md)

**Collaboration plane (ADR 0020):** seven-phase Harness HITL; specialized skills after `revalidationConfirmed`; timeout=deny.

**Phase 2 (ADR 0019):** evidence bundles (`governance:evidence`), cloud session ledger, strict VDR for T1+ — see `@hr-foundation-governance`.

Precedence: Human PO → PO checkpoint → counsel → Security merge bar.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.