thermograph/infra/.claude/skills/key-gaps/SKILL.md

102 lines
5.4 KiB
Markdown
Raw Normal View History

---
name: key-gaps
description: Audit which secret keys are missing, partially configured, or drifting across Thermograph environments (prod, beta, dev) — comparing the live /etc/thermograph.env on each box and/or the SOPS vault files (deploy/secrets/*.yaml). Use when asked about "key gaps", "missing keys/secrets", "what's not set", cross-environment secret drift, or as a pre-deploy / pre-cutover safety check. Reads key NAMES only, never values.
---
# Key-gap audit
Reports, per environment and across them, where secret **keys** are missing or
inconsistent — with special weight on the failure modes that bite silently here.
Run the deterministic script; don't eyeball it:
```
python3 .claude/skills/key-gaps/key_gaps.py <env>=<source> [<env>=<source> ...]
```
`<source>` is either a **SOPS file** (`deploy/secrets/prod.yaml`) or a **key-list
file** (one `KEY` per line, or `KEY=`/`KEY:` lines). Keys are readable in a SOPS file
even encrypted, so **no age key or decryption is needed**. Exit status is non-zero if
any CRITICAL or required gap is found (so it also works as a CI/pre-deploy gate).
## Audit the SOPS vault (offline, no SSH)
```
python3 .claude/skills/key-gaps/key_gaps.py \
prod=deploy/secrets/prod.yaml beta=deploy/secrets/beta.yaml
```
## Audit the LIVE boxes (read-only, key names only)
infra: split the estate into vps1/vps2, moving beta next to prod and dev onto vps1 Environments stop being machines. Until now each box WAS an environment -- "beta" named both a deploy target and a host, "the desktop" named both the operator's computer and the dev server -- so every path could assume one environment per host and hardcode /opt/thermograph, /etc/thermograph.env and ports 8137/8080. That assumption ends here: vps1 75.119.132.91 Forgejo, Grafana/Loki, the portfolio site, and DEV (own Postgres, mesh-only on 10.10.0.2:8137) vps2 169.58.46.181 PROD and BETA as two Swarm stacks sharing one TimescaleDB instance, plus Centralis, Postfix, backups desktop AI model hosting + flex Swarm capacity, no environment Nothing live has moved; the ordered cutover is in infra/deploy/RUNBOOK-vps1-vps2-cutover.md. deploy/env-topology.sh is the single source of truth: env -> host, checkout, branch, deploy mode, stack name, env file, LB ports, DB role/database, service prefix. THERMOGRAPH_ENV is the input, and on vps2 it is the only thing distinguishing a beta deploy from a prod one -- deploy.sh refuses to run when it disagrees with the checkout it was invoked from. The host-wide secrets-env and deploy-mode markers survive only as a fallback for a single-environment box. Beta gets its own stack file rather than an overlay (a merge cannot REMOVE the db service, and beta having no database of its own is the point). Its services are prefixed beta-*: Swarm registers a service's short name as a DNS alias on every network it joins, so two stacks both calling a service `web` on the shared data network would let prod's frontend resolve a beta task. Prod's stack, env and LB config are untouched. One Postgres, two databases with two roles: deploy/db/provision-env-db.sh creates thermograph_beta (NOSUPERUSER, owns only its own database, CONNECT revoked from PUBLIC) plus a <role>_ro for ad-hoc queries, and refuses to run for the environment that owns the instance -- doing so would demote prod's bootstrap superuser. Fixes that co-residency would otherwise have broken silently: - CI secrets are keyed by host (VPS1_SSH_*, VPS2_SSH_*). SSH_* meant "beta" and also "the Forgejo box" because those shared a machine; that conflation is what once had the prod backup dumping beta. - The nightly backup dumps BOTH application databases to separate off-box prefixes, and fails loudly on a missing one instead of skipping it. - Alloy stopped deriving the `host` label from the node -- on vps2 that would have filed every beta line as prod, feeding prod's alert rules with beta's traffic. It is now derived per source, with a new `node` label for the machine, and beta's log volume is mounted via a vps2-only overlay. - ops/dbq.sh, ops/iceberg.sh and the secrets seed scripts derived their SSH target and env-file path from the topology instead of hardcoding beta to 75.119.132.91 -- which is vps1's address now. - Dev's overlay binds DEV_BIND_ADDR (loopback by default, the mesh address on vps1) instead of 0.0.0.0: correct for a home LAN box, a public exposure of unreviewed branches on a VPS. Terraform's hosts variable is keyed by machine with a nested environments map, and main.tf flattens (host, environment) pairs so two environments on one box cannot share a checkout. Caddy's single reference config is split per host. Docs, onboarding and the runbooks are updated throughout, including the security rationales that co-residency makes false: separate SSH credentials no longer put a host boundary between beta and prod, and the boundary that remains is the database and the filesystem.
2026-07-25 22:01:29 +00:00
Gather the key names over SSH (never the values), then audit. Hosts per the
root `CLAUDE.md` topology:
- **vps2** (`169.58.46.181`) hosts BOTH prod and beta as separate Swarm
stacks — TWO live env files on the same box: `/etc/thermograph.env` (prod)
and `/etc/thermograph-beta.env` (beta). "The live env file per environment"
is no longer one path per host; it's one path per environment, and both
environments' files live on this one host. Get both in the same SSH round
trip, or two separate commands against the same host — never assume one
host means one file here.
- **vps1** (`75.119.132.91`) hosts dev, at the same `/etc/thermograph.env`
path (it's a different host, so no collision with prod's file of the same
name).
```sh
K=~/.ssh/thermograph_agent_ed25519
infra: split the estate into vps1/vps2, moving beta next to prod and dev onto vps1 Environments stop being machines. Until now each box WAS an environment -- "beta" named both a deploy target and a host, "the desktop" named both the operator's computer and the dev server -- so every path could assume one environment per host and hardcode /opt/thermograph, /etc/thermograph.env and ports 8137/8080. That assumption ends here: vps1 75.119.132.91 Forgejo, Grafana/Loki, the portfolio site, and DEV (own Postgres, mesh-only on 10.10.0.2:8137) vps2 169.58.46.181 PROD and BETA as two Swarm stacks sharing one TimescaleDB instance, plus Centralis, Postfix, backups desktop AI model hosting + flex Swarm capacity, no environment Nothing live has moved; the ordered cutover is in infra/deploy/RUNBOOK-vps1-vps2-cutover.md. deploy/env-topology.sh is the single source of truth: env -> host, checkout, branch, deploy mode, stack name, env file, LB ports, DB role/database, service prefix. THERMOGRAPH_ENV is the input, and on vps2 it is the only thing distinguishing a beta deploy from a prod one -- deploy.sh refuses to run when it disagrees with the checkout it was invoked from. The host-wide secrets-env and deploy-mode markers survive only as a fallback for a single-environment box. Beta gets its own stack file rather than an overlay (a merge cannot REMOVE the db service, and beta having no database of its own is the point). Its services are prefixed beta-*: Swarm registers a service's short name as a DNS alias on every network it joins, so two stacks both calling a service `web` on the shared data network would let prod's frontend resolve a beta task. Prod's stack, env and LB config are untouched. One Postgres, two databases with two roles: deploy/db/provision-env-db.sh creates thermograph_beta (NOSUPERUSER, owns only its own database, CONNECT revoked from PUBLIC) plus a <role>_ro for ad-hoc queries, and refuses to run for the environment that owns the instance -- doing so would demote prod's bootstrap superuser. Fixes that co-residency would otherwise have broken silently: - CI secrets are keyed by host (VPS1_SSH_*, VPS2_SSH_*). SSH_* meant "beta" and also "the Forgejo box" because those shared a machine; that conflation is what once had the prod backup dumping beta. - The nightly backup dumps BOTH application databases to separate off-box prefixes, and fails loudly on a missing one instead of skipping it. - Alloy stopped deriving the `host` label from the node -- on vps2 that would have filed every beta line as prod, feeding prod's alert rules with beta's traffic. It is now derived per source, with a new `node` label for the machine, and beta's log volume is mounted via a vps2-only overlay. - ops/dbq.sh, ops/iceberg.sh and the secrets seed scripts derived their SSH target and env-file path from the topology instead of hardcoding beta to 75.119.132.91 -- which is vps1's address now. - Dev's overlay binds DEV_BIND_ADDR (loopback by default, the mesh address on vps1) instead of 0.0.0.0: correct for a home LAN box, a public exposure of unreviewed branches on a VPS. Terraform's hosts variable is keyed by machine with a nested environments map, and main.tf flattens (host, environment) pairs so two environments on one box cannot share a checkout. Caddy's single reference config is split per host. Docs, onboarding and the runbooks are updated throughout, including the security rationales that co-residency makes false: separate SSH credentials no longer put a host boundary between beta and prod, and the boundary that remains is the database and the filesystem.
2026-07-25 22:01:29 +00:00
# sudo: the env files are root-owned (0640). On these boxes `agent` can often
# read them directly too, but sudo works uniformly regardless.
#
# The character class must allow DIGITS. This was `^[A-Z_]+=` until 2026-07-24,
# which silently skipped every key whose name contains a digit — in this estate
# that is all six *_S3_* keys. Prod audited as 26 keys against 32 real ones, and
# the audit reported the S3 credentials as missing from both hosts. They were
# present the whole time, and the phantom was independently "confirmed" twice by
# people re-running the same pattern. An audit that under-reports is worse than
# no audit: the invented finding sends someone to provision a credential that
# already exists, and it was briefly recorded as a root cause of a real bug.
# The match still stops at the `=`, so no value is ever read.
infra: split the estate into vps1/vps2, moving beta next to prod and dev onto vps1 Environments stop being machines. Until now each box WAS an environment -- "beta" named both a deploy target and a host, "the desktop" named both the operator's computer and the dev server -- so every path could assume one environment per host and hardcode /opt/thermograph, /etc/thermograph.env and ports 8137/8080. That assumption ends here: vps1 75.119.132.91 Forgejo, Grafana/Loki, the portfolio site, and DEV (own Postgres, mesh-only on 10.10.0.2:8137) vps2 169.58.46.181 PROD and BETA as two Swarm stacks sharing one TimescaleDB instance, plus Centralis, Postfix, backups desktop AI model hosting + flex Swarm capacity, no environment Nothing live has moved; the ordered cutover is in infra/deploy/RUNBOOK-vps1-vps2-cutover.md. deploy/env-topology.sh is the single source of truth: env -> host, checkout, branch, deploy mode, stack name, env file, LB ports, DB role/database, service prefix. THERMOGRAPH_ENV is the input, and on vps2 it is the only thing distinguishing a beta deploy from a prod one -- deploy.sh refuses to run when it disagrees with the checkout it was invoked from. The host-wide secrets-env and deploy-mode markers survive only as a fallback for a single-environment box. Beta gets its own stack file rather than an overlay (a merge cannot REMOVE the db service, and beta having no database of its own is the point). Its services are prefixed beta-*: Swarm registers a service's short name as a DNS alias on every network it joins, so two stacks both calling a service `web` on the shared data network would let prod's frontend resolve a beta task. Prod's stack, env and LB config are untouched. One Postgres, two databases with two roles: deploy/db/provision-env-db.sh creates thermograph_beta (NOSUPERUSER, owns only its own database, CONNECT revoked from PUBLIC) plus a <role>_ro for ad-hoc queries, and refuses to run for the environment that owns the instance -- doing so would demote prod's bootstrap superuser. Fixes that co-residency would otherwise have broken silently: - CI secrets are keyed by host (VPS1_SSH_*, VPS2_SSH_*). SSH_* meant "beta" and also "the Forgejo box" because those shared a machine; that conflation is what once had the prod backup dumping beta. - The nightly backup dumps BOTH application databases to separate off-box prefixes, and fails loudly on a missing one instead of skipping it. - Alloy stopped deriving the `host` label from the node -- on vps2 that would have filed every beta line as prod, feeding prod's alert rules with beta's traffic. It is now derived per source, with a new `node` label for the machine, and beta's log volume is mounted via a vps2-only overlay. - ops/dbq.sh, ops/iceberg.sh and the secrets seed scripts derived their SSH target and env-file path from the topology instead of hardcoding beta to 75.119.132.91 -- which is vps1's address now. - Dev's overlay binds DEV_BIND_ADDR (loopback by default, the mesh address on vps1) instead of 0.0.0.0: correct for a home LAN box, a public exposure of unreviewed branches on a VPS. Terraform's hosts variable is keyed by machine with a nested environments map, and main.tf flattens (host, environment) pairs so two environments on one box cannot share a checkout. Caddy's single reference config is split per host. Docs, onboarding and the runbooks are updated throughout, including the security rationales that co-residency makes false: separate SSH credentials no longer put a host boundary between beta and prod, and the boundary that remains is the database and the filesystem.
2026-07-25 22:01:29 +00:00
ssh -i $K agent@169.58.46.181 'sudo grep -oE "^[A-Z][A-Z0-9_]*=" /etc/thermograph.env' > /tmp/prod.keys # vps2: prod
ssh -i $K agent@169.58.46.181 'sudo grep -oE "^[A-Z][A-Z0-9_]*=" /etc/thermograph-beta.env' > /tmp/beta.keys # vps2: beta
ssh -i $K agent@75.119.132.91 'sudo grep -oE "^[A-Z][A-Z0-9_]*=" /etc/thermograph.env' > /tmp/dev.keys # vps1: dev
python3 .claude/skills/key-gaps/key_gaps.py prod=/tmp/prod.keys beta=/tmp/beta.keys dev=/tmp/dev.keys
```
## Verify a SOPS cutover matches live
Before flipping a host onto the vault, confirm the rendered set equals the live set —
mix the two source types for the same box:
```
python3 .claude/skills/key-gaps/key_gaps.py live=/tmp/prod.keys vault=deploy/secrets/prod.yaml
```
The `Cross-environment drift` section should report **none** for tracked keys.
## Reading the output
- **CRITICAL** — a required, *self-generating* secret is missing (`THERMOGRAPH_AUTH_SECRET`,
`THERMOGRAPH_VAPID_PRIVATE_KEY/_PUBLIC_KEY`). Missing here is the worst case: the app
silently mints a new value on boot, invalidating every login session and push
subscription. Fix before deploying.
- **MISSING** — a required secret is absent (`POSTGRES_PASSWORD`, `REGISTRY_TOKEN`).
- **Feature GAP** — a feature is *partially* configured (e.g. `discord-account-linking`
has the app id but not the client secret), so it's silently broken. A fully-unset
feature is "off", not a gap.
- **Dependent-key GAP** — an optional key is set but the *other* key it needs to do
anything isn't (e.g. `THERMOGRAPH_DISCORD_WEATHER_CHANNEL` set with no
`THERMOGRAPH_DISCORD_BOT_TOKEN` — discord.py gates the post on both). Asymmetric,
unlike a feature group: the prerequisite key is fine set alone (the bot token alone
already enables DMs).
- **drift** — a tracked key is in some environments but not others; often intentional,
but worth a glance.
## Keeping it current
The secret manifest (which keys are required / self-generating / grouped into features /
dependent on another key) lives at the top of `key_gaps.py` in `REQUIRED`,
`OPTIONAL_SELF_GEN`, `FEATURE_GROUPS`, `DEPENDENT_OPTIONAL`, and `STANDALONE_OPTIONAL`.
When a new `THERMOGRAPH_*` credential is added to `deploy/thermograph.env.example` /
`deploy/secrets/`, add it there too — check the code path that reads it (`os.environ.get`
call site) to see whether it's truly required, self-generating, part of an all-or-nothing
feature, or only meaningful alongside another key, rather than guessing from the name.