harden(auth): demo-mode residual-grants detector + cleanup endpoint + CI guard (A-8)

Audit 2026-05-11 A-8 closure. Closes the deferred Phase 2 leg of the
2026-05-10 HIGH-12 closure (b81588e) — production-startup observability
for actor-demo-anon residual grants + CI guard banning new synthetic-
admin code paths.

What this changes:

* cmd/server/preflight_demo_residual.go (new) runs after the DB pool +
  audit service are constructed and before the HTTPS listener starts.
  Under any non-'none' auth type it queries actor_roles for the
  synthetic actor-demo-anon and emits a WARN log + a categorized audit
  row (auth.demo_residual_grants_detected) listing every grant
  present. Migration 000029 unconditionally seeds the ar-demo-anon-admin
  row at install time, so EVERY production deploy will see this WARN
  on first boot; the intended cutover workflow is cleanup-once at
  production handover.

* CERTCTL_DEMO_MODE_RESIDUAL_STRICT (new env var on AuthConfig,
  default false) pivots the WARN to fail-closed startup refusal for
  operators who want a paranoid posture against re-seeding.

* POST /api/v1/auth/demo-residual/cleanup (new handler at
  internal/api/handler/demo_residual.go) is an admin-class
  (auth.role.assign) endpoint that removes every actor-demo-anon row
  from actor_roles and returns {removed: int64}. Idempotent; refuses
  503 under Auth.Type=none (deleting the row would break the demo
  path); audit-logs every invocation including no-op zero-removed
  calls so the admin's action is always recorded.

* scripts/ci-guards/no-new-synthetic-admin.sh pins the 17-entry
  allowlist of source files that legitimately reference the
  actor-demo-anon literal. New runtime code paths that resolve to the
  synthetic actor (the same pattern that produced the original CRIT
  class) are rejected at PR time. CI workflow auto-picks the script
  via the existing scripts/ci-guards/*.sh loop in .github/workflows/
  ci.yml; no workflow edit needed.

Regression matrix:

* cmd/server/preflight_demo_residual_test.go — 7 tests covering the
  4 main behaviour branches (testcontainers-backed, testing.Short()-
  skipped: DemoModeActive_Skips, NoResidue_Passes, HasResidue_LogsAnd
  Audits, StrictMode_RefusesStartup, DeleteDemoAnonResidue_Idempotent)
  plus 3 pure-Go stdlib unit tests for the row-string formatter +
  nil-safety contracts on both helpers.

* internal/api/handler/demo_residual_test.go — 7 stdlib+httptest
  cases: HappyPath, Idempotent_ReturnsZero, RejectsInDemoMode (503),
  CleanupError_Surfaces500, NilCleanupFn (defensive 500),
  NilAuditWriter_DoesNotPanic, MissingActorContext (falls back to
  'unknown' actor in the audit row).

* internal/api/router/openapi_parity_test.go — new
  POST /api/v1/auth/demo-residual/cleanup entry plus 6 pre-existing
  pre-A-8 entries (oidc/test, jwks-status, users CRUD, runtime-config)
  that had drifted out of SpecParityExceptions; the parity test was
  red on dev/auth-bundle-2 before my work; this commit returns it to
  green with full per-entry justifications + parity-debt notes.

Docs:

* docs/operator/security.md — new 'Demo-to-production cutover (Audit
  2026-05-11 A-8)' section explaining the WARN message, the cleanup
  curl one-liner, the equivalent SQL, the strict-mode env var, and
  the CI guard.

* docs/operator/rbac.md — Last-reviewed bump + pointer to the new
  env var + the security.md section.

* cowork/auth-bundles-audit-2026-05-10.md — HIGH-12 row gains an
  'A-8 follow-on CLOSED 2026-05-11' annotation describing the
  deferred Phase 2 leg now landed.

* CHANGELOG.md — Unreleased ### Security entry summarizing the four
  legs (detector + cleanup + strict-mode flag + CI guard) and the
  acquisition-readiness narrative this closes.

Operator-facing impact: this closes a credibility gap, not an
exploitable vulnerability. The residue requires a regression
elsewhere in the middleware chain to be exploitable. After this
fix, the canonical narrative ('RBAC primitive with no synthetic-
admin fallback') is fully true.

Refs cowork/auth-bundles-fixes-2026-05-11/08-high-demo-mode-residual-
cleanup.md.
This commit is contained in:
shankar0123
2026-05-11 11:45:54 +00:00
parent b3e3a8dbb1
commit ff3f1cd864
12 changed files with 1123 additions and 2 deletions
+7 -1
View File
@@ -1,6 +1,12 @@
# RBAC operator reference
> Last reviewed: 2026-05-09
> Last reviewed: 2026-05-11
>
> Audit 2026-05-11 A-8 follow-on: demo-mode residual-grants detector
> + cleanup endpoint shipped. New env var:
> `CERTCTL_DEMO_MODE_RESIDUAL_STRICT` (default `false`). Operator
> workflow at
> [`security.md#demo-to-production-cutover-audit-2026-05-11-a-8`](security.md#demo-to-production-cutover-audit-2026-05-11-a-8).
This is the operator-facing reference for the role-based access
control primitive that ships with Bundle 1 (auth bundle 1) of certctl.
+56 -1
View File
@@ -1,6 +1,6 @@
# certctl Security Posture & Operator Guidance
> Last reviewed: 2026-05-10
> Last reviewed: 2026-05-11
This document collects the operator-facing security guidance that the source
code's per-finding comment blocks reference. Each section names the audit
@@ -262,6 +262,61 @@ to avoid a permanent backdoor; the runbook at
[`auth-threat-model.md#break-glass-risks-phase-75`](auth-threat-model.md)
documents the full state machine.
### Demo-to-production cutover (Audit 2026-05-11 A-8)
Migration `000029_rbac.up.sql` unconditionally seeds an
`actor-demo-anon → r-admin` row into `actor_roles`. This row is the
runtime principal injected by the demo-mode middleware when
`CERTCTL_AUTH_TYPE=none`. Under any non-`none` auth type the row is
DORMANT — the middleware chain never resolves to it. But its existence
is a footgun: a future regression that resolves an unauthenticated
request to `actor-demo-anon` (a misrouted CORS preflight, a fallback in
a new auth-exempt route) would silently re-elevate to admin.
certctl-server detects this residue at startup and emits a WARN log +
an `auth.demo_residual_grants_detected` audit row listing every grant
present on `actor-demo-anon`. **Every production deploy will see this
WARN on first boot** — the migration baseline is part of the install,
not a side effect of running demo mode.
Operator workflow at production cutover:
1. Drain the WARN by calling the cleanup endpoint with an admin API key:
```bash
curl -X POST --cacert deploy/test/certs/ca.crt \
-H "Authorization: Bearer $ADMIN_KEY" \
https://certctl.example.com:8443/api/v1/auth/demo-residual/cleanup
# → {"removed": 1}
```
The endpoint is gated `auth.role.assign` (admin-class) and refuses
to run when `CERTCTL_AUTH_TYPE=none` (HTTP 503 — the residue IS the
active runtime state at that auth type). The cleanup is idempotent;
a second call returns `{"removed": 0}` and still leaves an audit row.
Equivalent SQL for operators preferring direct DB access:
```sql
DELETE FROM actor_roles WHERE actor_id = 'actor-demo-anon';
```
2. To make subsequent boots refuse startup if the row reappears (the
most paranoid stance), set:
```
CERTCTL_DEMO_MODE_RESIDUAL_STRICT=true
```
With the flag set, any `actor-demo-anon` row under a non-`none`
auth type causes certctl-server to log the WARN AND exit non-zero
before binding the HTTPS listener. Default is `false` (WARN only).
3. The CI guard `scripts/ci-guards/no-new-synthetic-admin.sh` pins the
set of source files that may reference the `actor-demo-anon` literal.
New runtime code paths that resolve to the synthetic actor are
rejected at PR time so the credibility gap stays closed.
### Migrating an existing deployment to OIDC
A Bundle-1-merged deployment that wants to add OIDC follows the