# [DEPRECATED] - Use https://github.com/HumandDev/hu-ai-agent-plugin/tree/main/plugins/humand-product-workflow
# Humand Product Workflow

Shared Cursor skills and agent configuration for the Humand development team.

## Structure

```
.cursor/
  sprint-report-context.json
  teams.json
  scripts/
    fetch-jira-dev-info.sh
    fetch-jira-sprint-issues.sh
    generate-sprint-report.py
    process-sprint-data.py
    run-sprint-report.py
    search-prs-for-keys.sh
  skills/
    feature-estimate-plan/
      SKILL.md
    hu-team-staging-status/
      SKILL.md
      team-staging-status.py
    setup/
      SKILL.md
    sprint-report/
      SKILL.md
```

## Setup

### Local (developers)

1. Install and authenticate the [GitHub CLI](https://cli.github.com/): `gh auth login`
2. Set Jira API credentials: `export JIRA_EMAIL=you@humand.com` and `export JIRA_API_TOKEN=<token>` (get token from https://id.atlassian.com/manage-profile/security/api-tokens)
3. (Optional) Configure the Atlassian MCP server in Cursor (fallback for Jira, required for Confluence)
4. Configure the Notion MCP server in Cursor (`https://mcp.notion.com/mcp`) — skills publish output to Notion pages
5. Run `/setup` in a conversation to verify everything works

### Cloud Agents (product managers)

1. Open https://cursor.com/dashboard?tab=cloud-agents and go to **My Settings**.
2. Add secret `GH_TOKEN` (read-only):
   - In GitHub: **Settings -> Developer settings -> Personal access tokens -> Fine-grained tokens -> Generate new token**.
   - Minimum permissions: **Metadata: Read**, **Contents: Read**, **Pull requests: Read**, **Issues: Read**.
3. Add secret `JIRA_API_TOKEN` (read-only):
   - In Atlassian: **Account settings -> Security -> API tokens -> Create API token**.
   - Use an account with Jira browse/read permissions only.
4. Use Cloud Agents to run the skills:
   - Cursor Agents UI: https://cursor.com/agents
   - Slack: tag `@Cursor` with your request.

## Skills

| Skill | Purpose | Run |
|------|---------|-----|
| `/setup` | First-time setup wizard — verifies `gh`, Atlassian MCP, Notion MCP, repo access, and teams config | `/setup` |
| `/sprint-report <team>` | Cross-repo sprint health → Notion page: Jira tickets + linked PRs, status, blockers | `/sprint-report rhino` |
| `/feature-estimate-plan` | Evidence-backed per-repo effort assessment, feasibility, and execution plan → Notion page (output in Spanish) | Ask the agent to plan a feature |
| `/hu-team-staging-status` | Per-repo staging vs develop status → Notion page (Jira-first) | `/hu-team-staging-status shark` |

## Reusable Scripts

Scripts in `.cursor/scripts/` are shared across skills.

| Script | Purpose |
|--------|---------|
| `fetch-jira-sprint-issues.sh` | Fetch sprint issues via Jira REST API with automatic pagination |
| `fetch-jira-dev-info.sh` | Query Jira's dev-status REST API for linked PRs/branches per ticket |
| `get-team-config.py` | Resolve one team by key/alias/name or validate `teams.json` without printing the whole file |
| `process-sprint-data.py` | Merges MCP JSON pages, performs parent/subtask rollup, outputs compact digest + processed tickets JSON |
| `generate-sprint-report.py` | Takes Jira tickets JSON + optional PR data, categorizes tickets, outputs formatted markdown |
| `run-sprint-report.py` | End-to-end wrapper: resolves team, fetches Jira, searches PRs, generates report |
| `search-prs-for-keys.sh` | Batch-search PRs across all 6 repos for a set of Jira ticket keys |