operator-workstation.env

# AgentKeys operator-workstation env file — source this on YOUR LAPTOP.
#
# Companion to scripts/broker.env (which is for the broker host).
#
# Scope: shell vars used by AWS admin tooling + the demo walkthrough in
# docs/stage7-demo-and-verification.md (§0 prerequisites + §4 isolation
# proof + §16 live walkthrough). The broker process itself reads NONE
# of these — they exist for `aws s3 ls`, `aws sts assume-role-with-web-identity`,
# `scripts/inspect-inbound-email.sh`, and any other workstation-side
# admin command that needs to address the AWS account.
#
# Usage:
#   awsp agentkeys-admin                       # switch to the admin profile
#   set -a; source ./operator-workstation.env; set +a
#
# After sourcing, $BUCKET / $ACCOUNT_ID / $BROKER_HOST / $OIDC_ISSUER /
# $OIDC_PROVIDER_ARN / $REGION are all set, and the demo guide's bash
# blocks copy-paste cleanly.
#
# This file commits as-is — only the public account ID + role/bucket
# names live here. No secrets.

# AWS account that owns agentkeys-data-role + agentkeys-mail-* bucket
# (cloud-setup.md §3.1 / §3.2).
ACCOUNT_ID=429071895007

# Region for STS + S3.
REGION=us-east-1
# AWS_REGION mirrors $REGION so AWS SDK consumers (agentkeys CLI, boto3,
# aws-sdk-rust) pick up us-east-1 regardless of the local AWS_PROFILE's
# default region. Without this, AWS_PROFILE=agentkeys-admin (whose
# profile defaults to us-west-2 per CLAUDE.md "Per-profile default
# region is NOT uniform" trap) makes the agentkeys CLI's S3 GetObject
# resolve against us-west-2, producing a misleading "Backend
# unreachable: GetObject: service error" instead of NotFound.
# Set as an explicit alias so $REGION stays the single source of truth.
AWS_REGION=$REGION
AWS_DEFAULT_REGION=$REGION

# The broker's public hostname. Used for SSH targets, OIDC issuer
# byte-for-byte matching, and as the host for $OIDC_ISSUER.
BROKER_HOST=broker.litentry.org

# Parent DNS zone owning BROKER_HOST + MAIL_DOMAIN + the service-worker
# subdomains (audit.${ZONE}, signer.${ZONE}, …). Used by
# scripts/setup-cloud.sh + dns-upsert-workers.sh.
ZONE=litentry.org
# Route 53 hosted zone ID for $ZONE. Discover via:
#   aws route53 list-hosted-zones --query 'HostedZones[?Name==`'"$ZONE"'.`].Id' --output text
PARENT_ZONE_ID=Z09723983CFJOHAE3VC65

# S3 bucket holding inbound mail (cloud-setup.md §2.2). Used by the
# demo's S3 isolation proof and inspect-inbound-email.sh.
BUCKET=agentkeys-mail-${ACCOUNT_ID}

# OIDC issuer URL — must match the URL passed to
# `aws iam create-open-id-connect-provider --url` (cloud-setup.md §4.2)
# byte-for-byte. The broker's BROKER_OIDC_ISSUER on the broker host is
# this same string.
OIDC_ISSUER=https://${BROKER_HOST}

# IAM OIDC provider ARN, derived from $ACCOUNT_ID + $BROKER_HOST.
OIDC_PROVIDER_ARN=arn:aws:iam::${ACCOUNT_ID}:oidc-provider/${BROKER_HOST}

# Federated role ARN — used by the daemon-side
# `aws sts assume-role-with-web-identity` calls in the demo. The daemon
# fetches an OIDC JWT from /v1/mint-oidc-jwt and does
# AssumeRoleWithWebIdentity client-side (issue #71 Option A; issue #72
# retired the broker-side /v1/mint-aws-creds aggregator).
#
# Stage-1 v2 split per arch.md §17.2 (per-bucket IAM role):
# - DATA_ROLE_ARN   → email subsystem (inbound/sent paths). Legacy name
#                     kept until email-service migrates in stage 2.
# - VAULT_ROLE_ARN  → credentials subsystem (bots/<actor_omni>/credentials/*).
#                     Provisioned by scripts/provision-vault-role.sh.
# - MEMORY_ROLE_ARN → memory subsystem (bots/<actor_omni>/memory/*).
#                     Provisioned by scripts/provision-memory-role.sh
#                     (added in issue #90 Q3 follow-up).
DATA_ROLE_ARN=arn:aws:iam::${ACCOUNT_ID}:role/agentkeys-data-role
VAULT_ROLE_ARN=arn:aws:iam::${ACCOUNT_ID}:role/agentkeys-vault-role
MEMORY_ROLE_ARN=arn:aws:iam::${ACCOUNT_ID}:role/agentkeys-memory-role

# Dedicated per-data-class bucket for stored credentials per arch.md §17
# (creds + email MUST live in separate buckets; sharing collapses
# encryption/lifecycle/CloudTrail blast radii). Provisioned by
# scripts/provision-vault-bucket.sh. Used by `agentkeys store/read` via
# AGENTKEYS_BUCKET=$VAULT_BUCKET in the orchestrator. The mail bucket
# ($MAIL_BUCKET, below) is no longer used for credentials.
VAULT_BUCKET=agentkeys-vault-${ACCOUNT_ID}

# Dedicated bucket for long-term agent memory blobs per arch.md §17.2.
# Distinct from VAULT_BUCKET (credentials) — different blast radius,
# different lifecycle policy. Provisioned by scripts/provision-memory-bucket.sh.
MEMORY_BUCKET=agentkeys-memory-${ACCOUNT_ID}

# ─── Signer (dev_key_service, issue #74 step 1b) ─────────────────────────────
# The dedicated signer listener (`agentkeys-signer.service`, :8092 loopback)
# is fronted publicly by nginx at a separate hostname under the same parent
# zone as the broker. Convention: `signer.<zone>` where <zone> is the broker's
# parent (broker.litentry.org → signer.litentry.org).
#
# Co-located with the broker today — same EC2 host, same IP, same nginx,
# same systemd box. `setup-broker-host.sh` provisions both. The split into a
# separate hostname (vs path-on-broker) is what lets us migrate the signer
# to a different machine (or TEE worker) later without changing the public
# API: clients keep talking to `https://signer.<zone>`, only the A record
# moves. See `cloud-setup.md §1.3` (DNS+TLS) and `docs/spec/architecture.md`
# §1 + §10 for the deployment topology.
#
# Used by:
#   - agentkeys CLI: `agentkeys signer derive/sign --signer-url $AGENTKEYS_SIGNER_URL`
#   - agentkeys-daemon: `--signer-url $AGENTKEYS_SIGNER_URL`
#   - the demo walkthrough §0.2 / §3 / §6 in stage7-demo-and-verification.md
SIGNER_HOST=signer.${BROKER_HOST#*.}
AGENTKEYS_SIGNER_URL=https://${SIGNER_HOST}

# Legacy alias kept so older copy-paste blocks (BACKEND_URL) keep working.
# New code should reference $AGENTKEYS_SIGNER_URL directly.
BACKEND_URL=${AGENTKEYS_SIGNER_URL}

# ─── Service workers (dev co-location on the broker host, issue #90) ─────────
# All four service workers (audit / email / credentials / memory) live on
# the same EC2 box as the broker today — co-location is dev-only per
# CLAUDE.md ("for production, we will isolate all the services for the
# security issue"). The per-worker hostnames are the migration seam: when
# a worker moves to its own machine, only the A record changes.
#
# `setup-broker-host.sh` provisions all four nginx vhosts + systemd units
# on the broker host. Operator laptop only needs the URLs for CLI tooling
# (e.g. `agentkeys audit query …` → $AGENTKEYS_WORKER_AUDIT_URL).
WORKER_AUDIT_HOST=audit.${BROKER_HOST#*.}
WORKER_EMAIL_HOST=email.${BROKER_HOST#*.}
WORKER_CRED_HOST=cred.${BROKER_HOST#*.}
WORKER_MEMORY_HOST=memory.${BROKER_HOST#*.}
AGENTKEYS_WORKER_AUDIT_URL=https://${WORKER_AUDIT_HOST}
AGENTKEYS_WORKER_EMAIL_URL=https://${WORKER_EMAIL_HOST}
AGENTKEYS_WORKER_CRED_URL=https://${WORKER_CRED_HOST}
AGENTKEYS_WORKER_MEMORY_URL=https://${WORKER_MEMORY_HOST}

# ─── CLI session storage ─────────────────────────────────────────────────────
# Force the `agentkeys` CLI to read/write the session JWT in a regular file
# (`~/.agentkeys/master/session.json`) instead of the macOS Keychain. Without
# this the CLI defaults to `KeyringMode::Auto` (per
# crates/agentkeys-core/src/session_store.rs:86), which:
#   1. Prompts for keychain access on every read (interactive blocker in
#      automated demo scripts; if the operator denies/dismisses, the CLI's
#      fallback path is non-obvious and can pick up a stale entry from
#      prior dev runs).
#   2. Returns `SIGNER_UNAUTHORIZED: invalid session JWT: InvalidToken`
#      from `agentkeys signer derive` if a stale Keychain entry exists,
#      even when `~/.agentkeys/master/session.json` has a fresh valid token.
#
# `file` mode keeps the demo Keychain-free end-to-end. To re-enable
# Keychain on a fresh machine, comment this line out and re-run
# `agentkeys init` — the CLI will write to the Keychain instead.
AGENTKEYS_SESSION_STORE=file

# ─── SES sender (Pass 1 of Option B — real email-link delivery) ──────────────
# Email subdomain — the SES domain identity verified per cloud-setup.md §1.1
# (DKIM/SPF/DMARC) AND the recipient root for the SES inbound receipt rule
# from §2.1 (any *@$MAIL_DOMAIN lands in s3://$BUCKET/inbound/). Distinct
# from $BROKER_HOST's zone — the operator may host the broker under a
# different parent domain than the email subdomain.
MAIL_DOMAIN=bots.litentry.org
MAIL_BUCKET=agentkeys-mail-${ACCOUNT_ID}

# The verified SES per-address identity the broker (and the integration test
# in crates/agentkeys-broker-server/tests/ses_email_flow.rs) uses as the FROM
# of magic-link emails. Must be registered + verified BEFORE first use:
#
#   bash scripts/ses-verify-sender.sh   # one-shot: create-identity → poll S3
#                                       # for verification mail → click link
#
# Same env var name (BROKER_EMAIL_FROM_ADDRESS) the broker reads at runtime
# (per crates/agentkeys-broker-server/src/env.rs:143). Setting it here means
# the test + the broker share one source of truth.
BROKER_EMAIL_FROM_ADDRESS=noreply-test@${MAIL_DOMAIN}
SCOPE_CONTRACT_ADDRESS_HEIMA_PASEO=0x0000000000000000000000000000000000000001
SIDECAR_REGISTRY_ADDRESS_HEIMA_PASEO=0x0000000000000000000000000000000000000002
K3_EPOCH_COUNTER_ADDRESS_HEIMA_PASEO=0x0000000000000000000000000000000000000003
CREDENTIAL_AUDIT_ADDRESS_HEIMA_PASEO=0x0000000000000000000000000000000000000004
HEIMA_PASEO_DEPLOYER_ADDR=0xeBdE9E5F8c0495e87a871BF4f17Fb85e1bFE827F
SCOPE_CONTRACT_ADDRESS_HEIMA=0xd44b375daefc65768f417d0f0125b68d5ba7df3b
SIDECAR_REGISTRY_ADDRESS_HEIMA=0x1Ac62f1C2D828476a5D784e850a700dC1f17e0bE
K3_EPOCH_COUNTER_ADDRESS_HEIMA=0x6c9e675c699a06acefbc156afdee6bfbfe32ccb3
CREDENTIAL_AUDIT_ADDRESS_HEIMA=0x63c4545ac01c77cc74044f25b8edea3880224577
HEIMA_DEPLOYER_ADDR_HEIMA=0xdE644936D5B7d5d42032fd08bbA42Fbbfd6663Bc
HEIMA_DEPLOYER_ADDR_HEIMA_PASEO=0xdE644936D5B7d5d42032fd08bbA42Fbbfd6663Bc
P256_VERIFIER_ADDRESS_HEIMA=0xda5b772f9d6c09abe80414eea908612df9b54749
K11_VERIFIER_ADDRESS_HEIMA=0x5a441431f08e0f5f5ed10659620cb4e0e814e627

# EC2 + EIP wiring lives in scripts/broker.env (the broker-machine env file)
# — those values identify the broker host, not operator-account identifiers.
# setup-cloud.sh sources broker.env after operator-workstation.env.