v1.6.1

Bug fixes found in role-based testing

1. AWS_PROFILE env var is now respected

Previously the CLI always read aws_profile from ~/.heyamara/config.json, so:

AWS_PROFILE=poweruser heyamara env pull ats-backend staging  # ← actually used 'backend' 🐞

Now the env var takes precedence. Config file is still the default when no env var is set.

2. heyamara db psql/url/run/doctor default to --as developer

Service users (ats_backend, etc.) don't have rds_iam grants — they're for K8s services only. Humans auth as developer. Pass --as <user> only when overriding.

# Old: explicit --as developer required
heyamara db psql staging ats --as developer

# New: same effect, no flag needed
heyamara db psql staging ats

3. Update-check only fires when remote is strictly newer

Fixes the 1.6.0 → 1.5.0 false positive.

4. kubectl auth errors classified correctly

Unauthorized from EKS no longer mis-reports as "service is not deployed in staging". You now get a clear "EKS access not granted" message with the fix.

Upgrade

heyamara update