> ## Documentation Index
> Fetch the complete documentation index at: https://docs.aegismemory.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Production Checklist

> What to configure before running Aegis Memory in production

# Production Checklist

A checklist for self-hosting Aegis Memory. All settings are environment variables; see
`.env.example` for the full annotated list.

## Secrets & authentication

* [ ] **`AEGIS_API_KEY`** — set a strong, unique API key. Clients authenticate with
  `Authorization: Bearer <key>`.
* [ ] **`AEGIS_INTEGRITY_KEY`** — set a dedicated HMAC signing key for memory integrity.
  If unset it falls back to `AEGIS_API_KEY`; a separate key is recommended so
  rotating one does not invalidate the other.
* [ ] **`OPENAI_API_KEY`** — required for embeddings unless you run an alternate
  embedding backend.
* [ ] Store all secrets in your platform's secret manager, not in the image or repo.

## Database

* [ ] PostgreSQL with the **`pgvector`** extension available to the app role.
* [ ] **`POSTGRES_PASSWORD`** set; database reachable from the API.
* [ ] Run migrations before first start: `alembic upgrade head`
  (see [Migrations](/deployment/migrations)).
* [ ] Tune the connection pool for your load: `DB_POOL_SIZE`, `DB_MAX_OVERFLOW`.

## Content security

* [ ] **`ENABLE_INTEGRITY_CHECK=true`** (default) — HMAC tamper detection on store/verify.
* [ ] Review content policy actions (`reject | redact | flag | allow`):
  `CONTENT_POLICY_PII`, `CONTENT_POLICY_SECRETS`, `CONTENT_POLICY_INJECTION`.
* [ ] Set limits appropriate to your data: `CONTENT_MAX_LENGTH`, `METADATA_MAX_DEPTH`,
  `METADATA_MAX_KEYS`.
* [ ] Decide whether to enforce trust levels: **`ENABLE_TRUST_LEVELS`** (default `false`).

## Rate limiting & quotas

* [ ] Global limits: `RATE_LIMIT_PER_MINUTE`, `RATE_LIMIT_PER_HOUR`, `RATE_LIMIT_BURST`.
* [ ] Per-agent limits: `PER_AGENT_RATE_LIMIT_PER_MINUTE`, `PER_AGENT_RATE_LIMIT_PER_HOUR`.
* [ ] Per-agent memory cap: `AGENT_MEMORY_LIMIT` (default `10000`).
* [ ] For multi-replica deployments, use Redis for distributed limiting: `REDIS_URL`.

## Networking

* [ ] **`CORS_ORIGINS`** — set explicit origins in production when sending
  `Authorization`/cookies. Use `*` only for public, non-credential APIs.
* [ ] Terminate TLS at your ingress/load balancer.

## Operations

* [ ] Configure backups and test a restore — see
  [Backup & Restore](/deployment/backup-restore).
* [ ] **Memory pruning is manual/CLI-triggered** in OSS (there is no hosted scheduled
  policy). Run `aegis prune` on your own schedule (e.g. a cron job) to archive stale
  memories via the temporal-decay sweep.
* [ ] Wire up observability — see [Observability](/guides/observability).

## Scope note

The open-source distribution is self-hosted and does **not** include managed backup
automation, a hosted dashboard, SSO/SAML/SCIM, org roles/billing, or external SIEM/
warehouse exporters. Plan those into your own infrastructure as needed.