em-intel — Engineering Manager Intelligence
Track team performance, engineer contributions, and project health across GitLab/GitHub + Jira/GitHub Issues.
Agent Instructions (read this first)
When this skill is triggered:
1. Check if configured:
test -f <skill_dir>/.env && echo "configured" || echo "not_configured"
If not configured, say:
"em-intel needs a one-time setup. Run this and I'll guide you through it (takes ~2 min):"
python3 <skill_dir>/em_intel.py setupOr to preview without any credentials first:
python3 <skill_dir>/em_intel.py morning-brief --dry-run
The setup command handles everything: asks questions, opens token pages in the browser, writes .env, installs deps, and runs doctor automatically.
If configured, run the requested command directly:
cd <skill_dir> && python3 em_intel.py <command> [--dry-run]
On error, run doctor and surface the failing checks:
python3 <skill_dir>/em_intel.py doctor
Quick Start (manual)
# 1. Copy and fill in your API keys
cp .env.example .env
# See SETUP.md for token URLs and required scopes
# 2. Install dependencies
pip3 install -r requirements.txt
# 3. Validate configuration
python3 em_intel.py doctor
# 4. Preview with mock data (no credentials needed)
python3 em_intel.py morning-brief --dry-run
# 5. Run for real
python3 em_intel.py morning-brief
Commands
| Command | Description |
|---|---|
doctor | Check env vars and test API connections |
morning-brief [--dry-run] | Merged yesterday, open PRs >3d, quiet engineers, stalled epics |
eod-review [--dry-run] | Today's merges/opens, contributor list, cycle time trend |
team-report [--days N] [--dry-run] | Full team performance report |
contributions [--engineer NAME] [--days N] [--dry-run] | Branch→ticket contribution map |
quiet-engineers [--dry-run] | Engineers with no MR activity |
epic-health [--dry-run] | Stalled and unassigned epics |
newsletter [--week] | Weekly digest via configured delivery channel |
--dry-run
Pass --dry-run to any supported command to use realistic synthetic mock data instead of hitting real APIs. Useful for previewing output format before configuring credentials.
Configuration
Set EM_CODE_PROVIDER to gitlab or github, and EM_TICKET_PROVIDER to jira or github_issues.
Delivery channels: telegram, slack, email, or print (stdout fallback).
See .env.example for all configuration options.
Architecture
em_intel.py ← CLI entrypoint (argparse)
adapters/ ← Code platform + ticket system adapters
base.py ← Abstract base classes & data models
gitlab_adapter.py ← GitLab REST API
github_adapter.py ← GitHub REST API
jira_adapter.py ← Jira REST API
github_issues_adapter.py ← GitHub Issues as ticket system
mock_adapter.py ← Synthetic data for --dry-run mode
core/ ← Business logic
branch_mapper.py ← Map branches → tickets → engineers
team_pulse.py ← Quiet engineers, MR trends, cycle times
jira_health.py ← Stale epics, unassigned tickets, PR age
newsletter.py ← Weekly digest generation
delivery.py ← Telegram / Slack / Email / Print routing
commands/ ← Command implementations
morning_brief.py ← Morning briefing
eod_review.py ← End-of-day review
team_report.py ← Full team report
newsletter.py ← Newsletter generation & delivery