Env validation falls back to localhost defaults in non-production, masking misconfiguration

[cybermax4200]

Why this matters now: A staging/preview deploy with a wrong DATABASE_URL (or missing JWT_SECRET) silently connects to localhost instead of failing fast, hiding outages and secret gaps.

Problem / What: validate.ts:88-118 — when safeParse fails and NODE_ENV !== 'production', it logs a warning and returns a fallback env with DATABASE_URL and JWT_SECRET defaults. So a misconfigured staging instance "starts successfully" against the wrong database and a known dev JWT secret.

Key Challenges: Defining which env vars are mandatory per environment (staging should fail closed on DATABASE_URL/secrets even if not "production"), without breaking local dev convenience.

Acceptance Criteria:

• A staging deployment missing/incorrect DATABASE_URL or JWT_SECRET fails fast with a clear error.
• Local development keeps ergonomic defaults (test proves both).

Relevant files/functions: src/config/validate.ts:84-118, src/config/startup.ts, src/config/jwtSecret.ts.

Out of scope: Moving config to a secrets manager.

Labels: advanced, ops, bug

[Divine-mercyx]

Hello, i would like to work on this issue

[grantfox-oss]

🦊 GrantFox — @Divine-mercyx has been assigned to this issue as part of the Third Campaign campaign!

Next steps:

1. Open a Pull Request referencing this issue (e.g., Closes #83)
2. Your PR will be reviewed by the ecotask-network maintainers

Good luck! Track your progress on GrantFox.