Commit 294f6cf (the prior docs fix for the multi-profile env vars)
introduced two doc-only env-var literals that the G-3 scanner picked
up as unmapped:
* CERTCTL_SCEP_PROFILE_CORP_ISSUER_ID — the literal CORP example
placeholder I added to clarify what the <NAME> substitution looks
like in practice. The G-3 scanner can't tell a placeholder from a
real env var.
* CERTCTL_SCEP_ — comes from the docs string CERTCTL_SCEP_* (the
asterisk is not in [A-Z_], so the regex strips it down to the
prefix and treats it as a phantom env var).
Two-part fix:
docs/features.md
* Replaced the literal CORP example (CERTCTL_SCEP_PROFILE_CORP_ISSUER_ID)
with a prose explanation that doesn't include a literal
placeholder env var name. Operators still get a clear example via
'a CERTCTL_SCEP_PROFILES entry of corp resolves the issuer-id env
var key with <NAME> replaced by CORP'.
.github/workflows/ci.yml
* Added CERTCTL_SCEP_ to the G-3 ALLOWED prefix list, mirroring the
existing CERTCTL_TLS_ entry. Both are legitimate doc-only prefix
references (CERTCTL_TLS_* / CERTCTL_SCEP_*) that the scanner sees
as bare prefixes after stripping the wildcard. The allowlist
documents these as integration-surface contracts that the
structured per-profile env vars expand into at runtime.
Verification: local G-3 set difference (Go-defined ∖ docs-mentioned)
empty in BOTH directions after the fix:
* DOCS_ONLY (docs ∖ Go, post-allowlist): empty
* CONFIG_ONLY (Go ∖ docs): empty
Restores green CI on the env-var docs drift guard.
Phase 1.5 added two new env-var literals to internal/config/config.go
that the G-3 docs-drift CI guard picked up but I forgot to document
when shipping commit fdd424b:
* CERTCTL_SCEP_PROFILES — comma-list of profile names enabling
multi-endpoint dispatch (e.g. 'corp,iot' produces /scep/corp +
/scep/iot).
* CERTCTL_SCEP_PROFILE_ — the prefix string used in
loadSCEPProfilesFromEnv's getEnv calls (e.g.
getEnv('CERTCTL_SCEP_PROFILE_'+envName+'_ISSUER_ID', ...)). The
G-3 regex extracts string literals between double quotes; the
prefix is a literal even though the suffix is concatenated at
runtime, so the scanner correctly flags it as 'defined in Go but
not documented'.
Added 7 rows to the SCEP env-vars table in docs/features.md:
* CERTCTL_SCEP_PROFILES (the explicit list var)
* CERTCTL_SCEP_PROFILE_<NAME>_ISSUER_ID (per-profile issuer)
* CERTCTL_SCEP_PROFILE_<NAME>_PROFILE_ID (per-profile cert profile)
* CERTCTL_SCEP_PROFILE_<NAME>_CHALLENGE_PASSWORD (per-profile secret)
* CERTCTL_SCEP_PROFILE_<NAME>_RA_CERT_PATH (per-profile RA cert)
* CERTCTL_SCEP_PROFILE_<NAME>_RA_KEY_PATH (per-profile RA key)
Each row notes the per-profile validation contract (required for every
profile in the list, file modes, fail-loud-with-PathID semantics).
Verification: local G-3 set difference (Go-defined ∖ docs-mentioned)
empty. The literal prefix CERTCTL_SCEP_PROFILE_ now appears in
docs/features.md as the documented env-var prefix, satisfying the
scanner's substring match.
SCEP RFC 8894 + Intune master bundle — Phase 0 + Phase 1 of 14.
Phase 0 (recon, no code changes):
Baseline tests green at HEAD 2519da8 (handler 79.0% / service 73.2% /
pkcs7 100%). SCEPConfig actual line is 666, prompt cited 639 — used
actual per the 'repo wins' operating rule.
Phase 1 (this commit):
internal/domain/scep.go
* Added SCEPMessageTypeCertRep (3) — RFC 8894 §3.3.2 server response
messageType. Clients pivot on this to extract a cert (Status=Success),
surface a failInfo (Status=Failure), or poll (Status=Pending).
* Added SCEPMessageTypeRenewalReq (17) — RFC 8894 §3.3.1.2
re-enrollment with an existing valid cert; signerInfo signed by the
existing cert (proving possession).
* Added SCEPRequestEnvelope struct — parsed authenticated attributes
from the inbound signerInfo (messageType / transactionID /
senderNonce / signerCert).
* Added SCEPResponseEnvelope struct — what the service hands back to
the handler so the handler can build the CertRep PKIMessage with
the correct status / failInfo / nonce echoes.
* Existing constants preserved unchanged.
internal/config/config.go
* SCEPConfig.RACertPath + RAKeyPath fields with the doc-comment density
matching the existing ChallengePassword field.
* Env-var loading: CERTCTL_SCEP_RA_CERT_PATH + CERTCTL_SCEP_RA_KEY_PATH.
* Validate() refuse: SCEP enabled with empty RA pair fails loud at
startup (defense-in-depth with the new preflight gate below).
cmd/server/main.go
* preflightSCEPRACertKey: file existence, mode 0600 gate (refuses
world-/group-readable RA key), tls.X509KeyPair-based parse + match
+ algorithm check (one stdlib call covers parse + cert-key match +
pubkey alg in one shot), expiry check, RSA-or-ECDSA gate (RFC 8894
§3.5.2 CMS signing requirement). Mirrors preflightSCEPChallenge-
Password's no-op-when-disabled pattern; each failure returns a
wrapped error so the caller (main) translates to a structured
slog.Error + os.Exit(1).
* Wired into the SCEP startup block immediately after the existing
challenge-password preflight; if it errors, the server refuses to
boot with a specific log line + the pointer to docs/legacy-est-scep.md
for the openssl recipe.
* Added crypto/tls + crypto/x509 imports.
cmd/server/preflight_scep_ra_test.go (new)
* Seven hermetic table-driven test cases covering each failure mode
spelled out in the helper's docblock plus the no-op-when-disabled
path. Each case materialises a real ECDSA P-256 cert/key pair on
disk so the tls.X509KeyPair path is exercised end-to-end (catches
drift in stdlib cert-parsing semantics that a mock would hide):
- disabled SCEP no-op
- missing paths (3 sub-cases: both empty, cert only, key only)
- world-readable key (chmod 0644)
- valid pair (happy path)
- expired cert (NotAfter in past)
- mismatched pair (cert from one ECDSA pair, key from another)
- missing files (paths set but files don't exist)
- ed25519 RA key (unsupported alg per RFC 8894 §3.5.2)
* writeECDSARAPair helper materialises a fresh ECDSA pair under the
test temp dir with the cert at 0644 and the key at 0600 (production
deploy mode).
internal/config/config_test.go
* TestValidate_SCEPEnabled_MissingRAPair_Refuses — 3 sub-cases pin
the new Validate() refuse path (both empty, cert only, key only).
* TestValidate_SCEPEnabled_CompleteRAPair_Accepts — pins the boundary
that file-existence is the preflight's job, NOT Validate's.
* TestValidate_SCEPDisabled_EmptyRAPair_Accepts — pins that the gate
only fires when SCEP is enabled (mirrors the CHALLENGE_PASSWORD
disabled-passes precedent).
docs/features.md
* SCEP env-vars table extended with CERTCTL_SCEP_RA_CERT_PATH and
CERTCTL_SCEP_RA_KEY_PATH (with the prod 'MUST set' callout +
file-mode 0600 requirement). Closes the G-3 'env var defined in Go
but never documented' CI guard for the new vars.
Verification:
* gofmt clean for the files I touched (preflight_scep_ra_test.go +
config.go + scep.go); pre-existing gofmt drift in unrelated files
not in scope.
* go vet ./internal/domain/... ./internal/config/... ./cmd/server/...
clean.
* go test -short -count=1 ./internal/domain/... ./internal/config/...
./cmd/server/... green.
* Coverage held at handler 79.0% / service 73.2% / pkcs7 100% /
config 96.1% / domain 88.6%.
* Local G-3 set difference (Go-defined env vars ∖ docs-mentioned env
vars) empty.
No behavior change for operators who don't enable SCEP. New behavior
gated by CERTCTL_SCEP_ENABLED=true + the new RA env vars. The MVP
raw-CSR fall-through path stays unchanged — Phase 2 will add the
RFC 8894 EnvelopedData decryption that consumes the RA pair.
Phase 1 of 14 in SCEP RFC 8894 + Intune master bundle.
Living progress at cowork/scep-rfc8894-intune/progress.md.
Audit pass against cowork/crl-ocsp-responder-prompt.md found three
operator-facing docs still describing the pre-bundle CRL/OCSP surface
(GET-only OCSP, CA-key-direct signing, no scheduler-driven cache). Each
claim updated below was ground-truthed against repo HEAD before edit.
README.md
* Standards & Revocation table — CRL row now mentions
scheduler-pre-generated cache (CERTCTL_CRL_GENERATION_INTERVAL,
crl_cache table); OCSP row mentions GET + POST forms, dedicated
responder cert per RFC 6960 §2.6, id-pkix-ocsp-nocheck per
§4.2.2.2.1, 7d auto-rotation grace.
* Revocation paragraph — corrected the 'Embedded OCSP responder'
one-liner to call out the dedicated-responder-cert design (the CA
private key is never used directly for OCSP signing, which is the
load-bearing security property for the future PKCS#11/HSM driver
path) and added the link to the relying-party guide.
docs/concepts.md
* CRL paragraph — added the scheduler pre-generation + singleflight
coalescing detail. Kept the existing 24h validity claim (verified
against internal/connector/issuer/local/local.go:956 — 'NextUpdate:
now.Add(24 * time.Hour)').
* OCSP paragraph — corrected the description so it covers both GET
and POST forms (POST per RFC 6960 §A.1.1 is what production
clients use: Firefox, OpenSSL s_client -status, cert-manager,
Intune); added the dedicated-responder-cert + nocheck-extension +
auto-rotation explanation; cross-link to docs/crl-ocsp.md.
docs/features.md
* Revocation Infrastructure section — CRL Endpoint, OCSP Responder,
new Admin Cache Observability subsection, new GUI Revocation
Endpoints Panel subsection. Corrected the previously-wrong 'Signs
with the issuing CA key' OCSP claim — the bundle's load-bearing
security improvement is exactly that the CA key is NOT used
directly. Cross-link to crl-ocsp.md.
* Local CA env vars table — added all four new
CERTCTL_CRL_GENERATION_INTERVAL / CERTCTL_OCSP_RESPONDER_KEY_DIR
(with the prod 'MUST set' callout) / _ROTATION_GRACE / _VALIDITY
rows. Closes the G-3 'env var defined in Go but never documented'
drift that broke CI on commit fc3c7ad.
* Migrations table — added 000019_crl_cache and 000020_ocsp_responder
rows so the table reflects the bundle's persisted surface area;
also clarified the table is illustrative + pointed readers at
'ls migrations/*.up.sql' for the full sequence (the table had
drifted behind reality at 000010 even before this bundle).
docs/architecture.md was already updated in commit b4334ed with the
same content scope, so no further architecture edits.
Verification:
* Local G-3 set difference: empty (Go-defined ∖ docs-mentioned for
CRL/OCSP env vars).
* 24h CRL validity claim verified against local.go:956 NextUpdate.
* Migration numbers verified against 'ls migrations/000019* 000020*'.
* id-pkix-ocsp-nocheck OID verified against
internal/connector/issuer/local/ocsp_responder.go:60.
Two CI failures on master after Bundle B merge:
1. Frontend Build / G-3 env-var docs guardrail
Bundle B introduced CERTCTL_RATE_LIMIT_PER_USER_RPS and
CERTCTL_RATE_LIMIT_PER_USER_BURST without adding them to
docs/features.md. The guardrail step that scans Go source for
getEnv* calls and asserts each appears in a doc page failed.
Fix: docs/features.md rate-limit section extended with both new
env vars + a paragraph explaining the per-key keying contract
from M-025.
2. Go Build & Test / staticcheck SA1019 hits (6 errors)
The CI workflow runs staticcheck without continue-on-error. Bundle
7 opened M-028 to track 6 deprecated-API sites; Bundle 9 closed 1
of them (the elliptic.Marshal in local.go) but kept a deliberate
regression-oracle reference in bundle9_coverage_test.go protected
only by golangci-lint's //nolint comment — staticcheck-as-CLI does
not honor that, only its native //lint:ignore directive.
Closure of remaining 5 sites:
cmd/server/main_test.go:47, 163, 192, 465 — 4 × middleware.NewAuth
migrated to middleware.NewAuthWithNamedKeys with explicit
NamedAPIKey entries. The auth=none case at line 465 maps to a
nil NamedAPIKey slice (no-op pass-through, matches the
NewAuthWithNamedKeys contract for empty input). Audit count was
3; recon found a 4th at line 465 that was missed.
internal/api/handler/scep.go:266 — csr.Attributes is a real RFC
2985 §5.4.1 challengePassword carve-out. Go's stdlib deprecation
note explicitly applies only to OID 1.2.840.113549.1.9.14
(requestedExtensions), NOT to OID 1.2.840.113549.1.9.7
(challengePassword), for which there is no non-deprecated
stdlib API. Suppressed with native //lint:ignore SA1019 +
comment block citing the RFC.
internal/connector/issuer/local/bundle9_coverage_test.go:342 —
deliberate regression-oracle that calls elliptic.Marshal to
prove the new crypto/ecdh path is byte-identical. Comment
converted from //nolint:staticcheck to native //lint:ignore
SA1019 so staticcheck-as-CLI honors the suppression.
Audit deliverables:
cowork/comprehensive-audit-2026-04-25/audit-report.md: M-028 box
flipped [x]; score 30/55 -> 31/55 (Medium 12/27 -> 13/27).
cowork/comprehensive-audit-2026-04-25/findings.yaml: M-028 status
partial_closed -> closed with closure note.
Verification:
go test -count=1 -short ./cmd/server ./internal/api/handler
./internal/connector/issuer/local ./internal/api/middleware
./internal/config — all green.
staticcheck on each changed package — 0 SA1019 hits.
Bundle C had M-028 in scope; this CI-fix lift moves it forward so
master CI goes green immediately. Bundle C scope adjusts to remove
M-028 and focuses on M-006 / M-015 / M-016 / M-019 / M-020 plus the
M-007 / M-008 coverage gaps.
Two CI gate failures from the Bundle 5 push:
1. golangci-lint (unused) — agent_bootstrap.go declared
`var bootstrapWarnOnce sync.Once` but never called .Do(). The
one-shot WARN actually lives in cmd/server/main.go (per-process at
startup, not per-request) so the handler-side variable was dead code.
Dropped the var + sync import; left a comment explaining where the
WARN lives.
2. G-3 env-var docs guardrail — Bundle 5 added two new env vars
(CERTCTL_AGENT_BOOTSTRAP_TOKEN, CERTCTL_AUDIT_FLUSH_TIMEOUT_SECONDS)
but the G-3 closure CI step asserts every CERTCTL_* env defined in
internal/config/config.go is mentioned in docs/features.md. Added
three new sub-sections to docs/features.md after the Body Size
Limits block:
* Agent Bootstrap Token (H-007 contract + generation guidance)
* Graceful Shutdown Audit Flush (M-011 timeout knob)
* Liveness vs Readiness Probes (H-006 /health vs /ready table)
No production behaviour change; pure CI-gate fix.
Verification
- go vet ./internal/api/handler/... → clean
- go test -count=1 -run 'TestVerifyBootstrapToken|TestRegisterAgent_BootstrapToken' ./internal/api/handler/... → all pass
- grep CERTCTL_AGENT_BOOTSTRAP_TOKEN docs/features.md → present
- grep CERTCTL_AUDIT_FLUSH_TIMEOUT_SECONDS docs/features.md → present
CI on the S-2 merge (a54805c) failed at the G-3 env-var-docs-drift
guardrail step:
G-3 regression: env var(s) defined in Go source but never documented:
CERTCTL_SHORT_LIVED_EXPIRY_CHECK_INTERVAL
The C-1 master commit (c4d231e) added the env var to
internal/config/config.go::SchedulerConfig + the Load() reader, and
wired the previously-dead Scheduler setter from cmd/server/main.go,
but I missed adding the env var to the canonical scheduler-loops
table at docs/features.md:1124.
Fix: the "Short-lived expiry check" row in the scheduler-loops table
now names CERTCTL_SHORT_LIVED_EXPIRY_CHECK_INTERVAL with the C-1
backstory ("pre-C-1 the setter was unwired and this env var had no
effect; post-C-1 it's read by cmd/server/main.go::sched.SetShortLived
ExpiryCheckInterval").
The G-3 guardrail is doing exactly what it was designed to do:
catching env-var docs drift the moment it appears. Working as
intended; this fix closes the gap the guardrail flagged.
Verification:
- comm -23 docs vs defined → empty post-fix (allowlist applied)
- comm -23 defined vs docs → empty post-fix
- The fix is doc-only; no Go / TS / config changes.
This is a follow-up to the C-1 + F-1 + P-1 + S-2 mega-prompt closure;
push together to unblock CI.
Closes six 2026-04-24 audit findings (3 P2 + 3 P3) — a cleanup-and-doc
tail bundle that drains the smallest remaining leaves of the audit:
- cat-u-vite_dev_proxy_plaintext_drift (P2): web/vite.config.ts
proxied dev requests to http://localhost:8443 against an HTTPS-only
backend (HTTPS-only since v2.0.47). Every dev-server API call 502'd.
Fix: targets are now object-form `{target: 'https://...', secure: false,
changeOrigin: true}` — the dev cert is self-signed by the
deploy/test bootstrap and changes per-checkout.
- cat-g-7e38f9708e20 (P3): Scheduler.SetShortLivedExpiryCheckInterval
was defined + tested but never called from cmd/server/main.go.
Operators tuning CERTCTL_SHORT_LIVED_EXPIRY_CHECK_INTERVAL got
no effect — the 30s default in scheduler.NewScheduler was
effectively hardcoded. Fix: added Config.Scheduler.ShortLivedExpiryCheckInterval
+ getEnvDuration in Load() reading the env var with a 30s default,
+ sched.SetShortLivedExpiryCheckInterval(...) call in main.go
alongside the other scheduler-interval setters.
- diff-10xmain-2bf4a0a60388 (P3): same root cause as cat-g-7e38f9708e20;
closes as ride-along.
- cat-b-6177f36636fb (P2): registerAgent client fn orphan. By-design
per pull-only deployment model. Fix (audit recommendation:
"document"): added a closure docblock above the export in
client.ts + a new "Registration is by-design pull-only" paragraph
in docs/architecture.md::Agents section explaining when/why a
future GUI-driven enrollment feature might reach the endpoint
(proxy-agent topologies for network appliances).
- cat-i-7c8b28936e3d (P2): CLI scope intentionally narrow but
undocumented. Fix: new "Scope (intentionally narrow)" subsection
in docs/features.md::CLI capturing the SSH-into-prod / day-to-day
GUI / AI-automation MCP three-way split.
Verification:
- go build ./... — clean
- go vet ./... — clean
- go test ./internal/scheduler/... ./internal/config/... — pass
- golangci-lint v2.11.4 run ./... — 0 issues
- tsc --noEmit (frontend) — clean
- All sibling guardrails (S-1 / G-3 / D-1+D-2 / B-1 / L-1 / H-1) still pass
Audit findings closed:
- cat-u-vite_dev_proxy_plaintext_drift (P2)
- cat-g-7e38f9708e20 (P3)
- diff-10xmain-2bf4a0a60388 (P3)
- cat-b-6177f36636fb (P2)
- cat-i-7c8b28936e3d (P2)
- (audit-bookkeeping ride-along: ensures every closed-bundle row has a non-empty merge SHA)
Deferred follow-ups: none from this bundle. The remaining audit
backlog (frontend test campaign, F-1 CertificatesPage UX, P-1
orphan-fn sweep, S-2 handler error-mapping refactor) is sibling
sub-bundles in this mega-prompt.
Closes three 2026-04-24 audit findings (all P2, all category cat-g):
- cat-g-renewal_check_interval_rename_drift: features.md:152
advertised CERTCTL_RENEWAL_CHECK_INTERVAL but config.go renamed
that to CERTCTL_SCHEDULER_RENEWAL_CHECK_INTERVAL. Fixed in prose
+ the scheduler-loops table on line 1117.
- cat-g-b8f8f8796159: 6 env vars in config.go that were never
documented:
CERTCTL_DATABASE_MIGRATIONS_PATH
CERTCTL_JOB_AWAITING_APPROVAL_TIMEOUT
CERTCTL_JOB_AWAITING_CSR_TIMEOUT
CERTCTL_SCHEDULER_AGENT_HEALTH_CHECK_INTERVAL
CERTCTL_SCHEDULER_JOB_PROCESSOR_INTERVAL
CERTCTL_SCHEDULER_NOTIFICATION_PROCESS_INTERVAL
Added to the scheduler-loops table at features.md:1117 and
(DATABASE_MIGRATIONS_PATH) to the new Database Schema preamble.
- cat-g-163dae19bc59: 37 env vars in docs not defined in config.go.
The audit's strict comm over-flagged this set: most "phantoms"
are integration-surface contracts (script env vars certctl
EXPORTS to user-provided ACME DNS-01 / OpenSSL CA scripts;
StepCA / Webhook per-issuer-or-notifier config-blob field
names; CERTCTL_QA_* test fixtures; agent-side env vars defined
in cmd/agent/main.go). The closure narrows the gate to the
one true phantom (the rename) and allowlists the documented
integration contracts in the CI guard. Each allowlist entry
has a one-line justification.
CI regression guardrail:
- .github/workflows/ci.yml::"Forbidden env-var docs drift regression
guard (G-3)" — runs `comm -23` both ways between the env vars
defined in Go source (config.go + cmd/* + ACME DNS export +
test fixtures) and env vars mentioned in README + docs/ +
deploy/helm/. Fails the build if either set is non-empty modulo
the documented integration-surface allowlist.
Verification:
- comm -23 docs vs defined → empty post-fix (allowlist applied)
- comm -23 defined vs docs → empty post-fix
- golangci-lint v2.11.4 run ./... → 0 issues
- tsc --noEmit → clean
- S-1 stale-counts guardrail still passes
Audit findings closed:
- cat-g-163dae19bc59 (P2, docs-only env vars)
- cat-g-b8f8f8796159 (P2, config-only env vars)
- cat-g-renewal_check_interval_rename_drift (P2, renamed env var still in docs)
Deferred follow-ups:
- The 26 documented-but-unimplemented integration contracts on the
allowlist (CERTCTL_OPENSSL_*, CERTCTL_ACME_EAB_*, CERTCTL_WEBHOOK_*,
CERTCTL_AUDIT_EXCLUDE_PATHS, CERTCTL_TLS_*, CERTCTL_ACME_DNS_PROPAGATION_WAIT)
are documented in features.md / connectors.md / demo-advanced.md but
not yet read by any Go source. Either implement in config.go (each is
its own M-X) or delete from docs (separate cleanup PR). Neither
expansion fits inside G-3's "reconcile drift" scope.
Closes two 2026-04-24 audit findings — one P1 (cat-s1-9ce1cbe26876,
README + features.md cite stale numeric counts) and one P2
(cat-s1-features_md_issuer_count_contradiction, features.md self-
disagreed on issuer count saying 9 in two places + 12 in two others).
Both root in a CLAUDE.md invariant: "Numeric claims about current
state rot the instant the next release lands... Before adding any
current-state count, delete it and write the command instead."
Per-site changes:
- docs/features.md::"At a Glance" table — replaced 12 hardcoded counts
with `rebuild via <command>` references quoting the canonical
source-of-truth grep from CLAUDE.md::"Current-state commands".
- docs/features.md::Issuer Connectors section — dropped "9 issuer
connectors" (stale; live: 12) and "12 IssuerType constants" prose;
prose now references the rebuild command.
- docs/features.md::Target Connectors section — same treatment for
"14 target connector types".
- docs/features.md::"Per-type config schema validation for all 9
issuer types" — same treatment.
- docs/features.md::"80 MCP tools covering all API endpoints" — same.
- docs/features.md::Web Dashboard section — dropped "24 pages wired"
+ the "(25 Route elements, 24 pages)" comment.
- docs/examples.md::"Beyond These Examples" — dropped "7 issuer
backends and 10 target connectors" prose; references features.md
and the rebuild commands.
CI regression guardrail:
- .github/workflows/ci.yml::"Forbidden hardcoded source-count prose
regression guard (S-1)" — grep-fails the build if any of the
blocked phrases (e.g. "9 issuer connectors", "21 database tables",
"80 MCP tools") reappears in README or docs/. Allowlists demo-
fixture prose ("32 certificates" — seed_demo.sql facts), historical
WORKSPACE-CHANGELOG counts, the testing-guide example phrasing,
and any number adjacent to a quoted rebuild command.
Verification:
- S-1 guardrail dry-run on post-fix tree → empty (good)
- golangci-lint v2.11.4 run ./... → 0 issues
- tsc --noEmit → clean
- vitest, vite build unchanged from pre-S-1 baseline (no JS/TS touched)
Audit findings closed:
- cat-s1-9ce1cbe26876 (P1, README + features.md stale numeric counts)
- cat-s1-features_md_issuer_count_contradiction (P2, features.md
self-contradiction on issuer count)
Deferred follow-ups:
- WORKSPACE-CHANGELOG.md historical-milestone counts intentionally
preserved (those are point-in-time facts about shipped slices, not
current-state claims). README demo-fixture counts ("32 certs, 10
issuers") preserved — those describe the seed_demo.sql shape, not
the live source surface.
Closes the remaining P1 gaps from coverage-gap-audit.md (M-001/M-002/M-003/M-006)
on top of the C-001/C-002 ownership + agent-FK contract fixes landed in
a53a4b8. The work lands as a single commit spanning server, docs, tests,
and the React client.
M-002 — Named API keys with per-key actor propagation
* Migration 000014 adds the 'api_keys' table (id, name, hash,
principal, role, created_at, last_used_at, disabled_at) so every
credential carries an identifiable principal instead of the
opaque 'anonymous'/'api-key' sentinel.
* Auth middleware now rotates through configured keys, performs
constant-time hash comparison, stamps 'last_used_at', and emits
an actor struct via contextWithActor(). The audit middleware,
bulk-revocation handler, approval handlers, and MCP tool layer
now read the principal off the context and persist it on every
audit_events row.
* Regression coverage:
- internal/api/middleware/audit_test.go — actor propagation,
principal redaction for disabled keys, anonymous fallback for
unauthenticated endpoints.
- internal/api/handler/bulk_revocation_handler_test.go,
job_handler_test.go — principal-on-audit assertions.
M-003 — Authorization gates (Phase B)
* Approval handler rejects self-approval / self-rejection with 403
when the actor principal equals the job's requested_by field.
* Bulk revocation is gated behind the 'admin' role; operators and
viewers receive 403.
* Regression coverage:
- internal/service/job_test.go — TestApproveJob_NotSelf,
TestRejectJob_NotSelf.
- internal/api/handler/bulk_revocation_handler_test.go —
TestBulkRevoke_RequiresAdmin, TestBulkRevoke_AdminSucceeds.
M-006 — RFC-compliant CRL/OCSP on the unauthenticated .well-known mux
* Per RFC 8615, relying parties cannot reasonably be asked to
authenticate against the issuing certctl instance to retrieve
revocation material. CRL and OCSP move off the authenticated
'/api/v1/crl*' and '/api/v1/ocsp/*' paths onto:
GET /.well-known/pki/crl/{issuer_id}
Content-Type: application/pkix-crl (RFC 5280 §5)
GET /.well-known/pki/ocsp/{issuer_id}/{serial}
Content-Type: application/ocsp-response (RFC 6960)
* Non-standard JSON CRL shape is removed; only DER is served.
* Short-lived certificate exemption (profile TTL < 1h → skip
CRL/OCSP) is preserved; the response simply omits the serial.
* Routes are registered on the unauthenticated 'finalHandler' mux
in cmd/server/main.go alongside EST ('/.well-known/est/*') and
SCEP ('/scep'). Legacy authenticated paths return 404.
* Regression coverage:
- internal/api/handler/certificate_handler_test.go — content
type, DER parseability, 404 for unknown issuer.
- internal/api/handler/adversarial_path_test.go — unauthenticated
access asserted for CRL, OCSP, EST, SCEP.
- internal/api/router/router_test.go — route-table assertion
that '.well-known/pki/*', '.well-known/est/*', and '/scep' are
mounted on the unauthenticated branch.
M-001 — Auto-closed by M-002
EST and SCEP were already registered on the unauthenticated
'finalHandler' mux; the router comment at
internal/api/router/router.go:247 now matches reality. The
adversarial-path tests above lock the behavior in.
Verification (all gates green):
* go vet ./... — clean
* go build ./... — ok
* go test -short ./... (55+ packages) — all pass
* web/ : npm test (225 Vitest tests) — all pass
* web/ : npx tsc --noEmit — clean
* grep sweep for '/api/v1/(crl|ocsp)' — 13 surviving hits,
all intentional M-006 tombstone/relocation comments.
Documentation:
* coverage-gap-audit.md — status flips M-001/M-002/M-003/M-006 →
Fixed, with per-finding resolution paragraphs citing regression
test IDs. (Audit file lives outside this repo; see cowork root.)
* CLAUDE.md Project Status line updated with the auth-unification
closure note.
* docs/features.md, docs/architecture.md, docs/quickstart.md,
docs/concepts.md, docs/connectors.md, docs/test-env.md,
docs/testing-guide.md, docs/compliance-*.md, docs/demo-advanced.md
— refreshed for the new '.well-known/pki/*' namespace and named
API keys.
* api/openapi.yaml — documents the new unauthenticated endpoints
and removes the legacy '/api/v1/crl*' + '/api/v1/ocsp/*' paths.
.gitignore: adds '/.gocache/' and '/.gomodcache/' for the session-
scoped Go caches so they never enter the tree.
Add three new issuer connectors completing commercial and open-source CA
coverage. Entrust uses mTLS client certificate auth with sync/async
issuance. GlobalSign Atlas uses mTLS + API key/secret dual auth with
serial-based tracking. EJBCA supports dual auth (mTLS or OAuth2) for
self-hosted Keyfactor CAs.
Each connector implements the full issuer.Connector interface (9 methods),
includes httptest-based unit tests (~14 each), and follows established
patterns (injectable HTTP clients, RFC 5280 revocation reason mapping,
CRL/OCSP delegated to CA).
Also includes: issuer factory cases, env var seeding, config structs,
domain types, seed data (3 rows, all disabled), OpenAPI enum updates,
frontend issuer catalog entries with config fields, and full docs
(connectors.md, architecture.md, features.md, README).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Enforce certificate profile crypto constraints across all 5 issuance paths
(renewal, agent CSR, EST, SCEP). ValidateCSRAgainstProfile() rejects CSRs
with key algorithm/size that don't match profile rules. MaxTTL enforcement
caps certificate validity per issuer connector (Local CA, Vault, step-ca
enforce directly; ACME/DigiCert/Sectigo pass through). Key algorithm and
size are now persisted in certificate_versions for audit compliance.
16 new tests (12 service-layer + 4 Local CA connector). Removes hardcoded
version number from GUI sidebar. Documentation updated across architecture,
features, connectors, and README.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Implements Simple Certificate Enrollment Protocol with single-endpoint
operation-based dispatch (GetCACaps, GetCACert, PKIOperation), PKCS#7
SignedData CSR extraction with fallback for raw/base64 CSR, challenge
password authentication via CSR attributes, and shared internal/pkcs7
package extracted from EST handler to eliminate code duplication.
24 new tests (11 service + 13 handler) plus 5 shared pkcs7 package tests.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Rewrote docs/features.md from scratch as authoritative feature inventory
(1255 lines, every claim verified against source files).
Audited README.md and architecture.md against repo — fixed 19 stale
references: K8s Secrets status, issuer counts, dashboard page counts,
CI thresholds, missing connectors in Mermaid diagrams, OpenAPI operation
count, GetCACertPEM behavior, and V2/V4 roadmap accuracy.
Also includes related fixes discovered during audit:
- Scheduler skips expired/failed/revoked certs from auto-renewal
- Seed demo expiry dates moved outside 31-day scheduler query window
- Agent pages use correct last_heartbeat_at field name
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Three related ACME ecosystem changes shipped as a single milestone:
1. ACME Certificate Profile Selection: Custom JWS-signed newOrder POST with
`profile` field (e.g., `tlsserver`, `shortlived` for 6-day certs) bypassing
acme.Client.AuthorizeOrder() since golang.org/x/crypto lacks profile support.
ES256 JWS signing with kid mode, nonce management, directory discovery.
Empty profile delegates to standard library path (zero behavior change).
Configurable via CERTCTL_ACME_PROFILE env var. GUI: profile dropdown on
ACME issuer config.
2. ARI RFC 9702 → 9773 Renumber: All 25+ references updated across Go source,
docs, README, and examples. Zero remaining occurrences of RFC 9702.
3. 45-Day / Short-Lived Certificate Positioning: 5 domain tests validating
renewal thresholds against SC-081v3 validity reduction timeline (200→100→47
days) and Let's Encrypt 45-day/6-day profiles. ARI (RFC 9773) is the
expected renewal path for 6-day shortlived certs.
New tests: 13 profile + 5 domain threshold + 1 frontend = 19 new tests.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace brittle test count numbers (1,554+, 1,088+, 211, etc.) with
descriptions of testing approach and CI-enforced coverage gates.
Counts go stale every milestone — coverage thresholds are machine-
verified and never drift.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Audit all docs and examples against current codebase state. Fix seed_demo.sql
domain constant casing (IssuerType, TargetType, AgentStatus) that would cause
agent dispatch failures. Fix example docker-compose health endpoints (/health
not /api/v1/health) and env var names (CERTCTL_DATABASE_URL). Update connector
counts, test numbers, and planned→implemented status across docs. Convert 3
ASCII flow diagrams to Mermaid.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Frontend audit (10 categories): lifecycle fields in types, new API
functions (CRL, OCSP, deployments, updateIssuer/Target, getPolicy),
issuer/owner/profile filters on CertificatesPage, last_renewal_at
column, error_message column on JobsPage, full crypto policy UI on
ProfilesPage (key algorithms, EKUs, SAN patterns), key info + CA
badge on DiscoveryPage, edit modal on TargetDetailPage, tags field
on certificate creation, darwin→macOS mapping on AgentFleetPage.
211 Vitest tests passing.
README accuracy: test counts (1300+ Go, 211 frontend), page count
(24), demo data (32 certs, 7 issuers, 180 days), endpoint count
(97), MCP tools (80), CLI subcommands (10), moved shipped items
out of "Coming in v2.1.0".
Docs: architecture.md diagrams updated (Vault PKI, DigiCert,
Traefik, Caddy added), features.md Vault/DigiCert status updated.
Version bumped to v2.0.20. cli binary removed from git tracking.
Testing guide Part 41 added (12 auto + 9 manual tests).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add certificate export in PEM (JSON or file download) and PKCS#12 formats.
Private keys are never included — they stay on agents. Add EKU-aware
issuance threading profile EKUs (serverAuth, clientAuth, codeSigning,
emailProtection, timeStamping) through the full issuance pipeline. Fix
agent CSR SAN splitting for email addresses, adaptive KeyUsage flags for
S/MIME vs TLS, and a pre-existing generateID collision bug in deployment
job creation.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Update README, architecture guide, and feature inventory to document all
changes from the security remediation pass (17 tickets):
- README: Add CI pipeline section (race detection, golangci-lint,
govulncheck, per-layer coverage thresholds), CORS deny-by-default
behavior, input validation, SSRF protection, scheduler concurrency
safety. Update test count to 1050+. Add race detection and govulncheck
to development commands.
- Architecture guide: Update testing strategy with scheduler tests, fuzz
tests, and revised CI pipeline description. Add security model sections
for input validation, CORS, and concurrency safety. Update test count.
- Feature inventory: Document CORS deny-by-default behavior.
- SECURITY_REMEDIATION.md: New file documenting all 17 remediated tickets
with CWE classifications, before/after behavior, 3 deferred tickets
with rationale, CI pipeline changes, and breaking CORS change.
Missing docs flagged as future additions:
- Formal threat model document
- Disaster recovery runbook
- Version upgrade guide
- Capacity planning benchmarks
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
M25: After deploying a certificate, the agent probes the live TLS
endpoint and compares SHA-256 fingerprints to verify the correct cert
is being served. Best-effort — failures don't block deployments.
New endpoints: POST /jobs/{id}/verify, GET /jobs/{id}/verification.
Migration 000008 adds verification columns to jobs table.
M26: Traefik target connector (file provider, auto-reload) and Caddy
target connector (dual-mode: admin API hot-reload or file-based).
Both wired into agent dispatch.
Also: restructured README to highlight supported integrations (issuers,
targets, notifiers) earlier, moved API/CLI/MCP sections lower. Updated
all docs (features, connectors, architecture, testing guide, why-certctl)
and fixed integration tests for 18-param RegisterHandlers signature.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Standing TXT record at _validation-persist.<domain> eliminates per-renewal
DNS updates. Auto-fallback to dns-01 if CA doesn't offer dns-persist-01.
ScriptDNSSolver extended with PresentPersist method. Configurable via
CERTCTL_ACME_CHALLENGE_TYPE=dns-persist-01 and
CERTCTL_ACME_DNS_PERSIST_ISSUER_DOMAIN env vars.
Also fixes IsExpired edge-case test in discovery_test.go that always failed
due to time.Now() drift between test setup and method invocation.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Implement Enrollment over Secure Transport protocol with 4 endpoints under
/.well-known/est/ — cacerts (CA chain distribution), simpleenroll (initial
enrollment), simplereenroll (certificate renewal), and csrattrs (CSR
attributes). PKCS#7 certs-only wire format with hand-rolled ASN.1, accepts
both PEM and base64-encoded DER CSRs, configurable issuer and profile
binding, full audit trail. 28 new tests (18 handler + 10 service).
Also includes:
- GetCACertPEM added to issuer connector interface (all 4 issuers updated)
- EST integration tests wired into e2e test suite (13 test cases)
- QA testing guide Part 26 (15 manual EST test cases)
- All docs updated: README, features, architecture, concepts, connectors,
quickstart, demo-advanced (endpoint counts, MCP wording, agent IDs,
issuer interface, resource lists, OpenSSL status)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Frontend: fetchJSON now returns empty object on 204 instead of failing
to parse empty body — fixes silent delete failures across all entities.
Added onError callbacks to owner/team delete mutations to surface errors.
Backend: owner and issuer delete handlers return 409 Conflict with
descriptive messages when FK constraints block deletion, instead of
generic 500.
Added 15 v2 dashboard screenshots, updated README screenshot section,
logo asset, page count references (18→full), and QA guide with FK
constraint test coverage.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
CLI syntax corrected across 5 files (concepts, demo-guide, demo-advanced,
architecture, features): list-certs→certs list, get-cert→certs get, etc.
Removed non-existent health/metrics commands, replaced with status.
Subcommand count 10→12 everywhere.
architecture.md: Go 1.22→1.25, endpoint count 91→93, ER diagram expanded
from 15 to 21 tables (added renewal_policies, certificate_revocations,
discovered_certificates, discovery_scans, network_scan_targets).
connectors.md: added GenerateCRL and SignOCSPResponse to issuer interface,
added Email and Webhook rows to notifier config table.
compliance docs: fixed keygen warning messages to match actual log output,
CERTCTL_STEPCA_PROVISIONER_KEY→CERTCTL_STEPCA_KEY_PATH, openssl genrsa→
crypto/ecdsa.GenerateKey, CERTCTL_SERVER_ADDR→CERTCTL_SERVER_HOST+PORT.
README.md: v2.0.0 version bump, solo developer mention, feature list,
table of contents, documentation table moved to top, 7 fact-check fixes.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
V2 audit found 3 critical number mismatches propagated across 8 files:
- Table count was 19 everywhere but actual migrations create 21 tables
- MCP tool count was 76 but tools.go registers 78 (M21/M22 additions)
- README MCP breakdown claimed 83 tools with math summing to 90
- architecture.md still had stale 860+ test count
- features.md OpenAPI claim said 93 ops but spec has 78
- mcp.md tool-per-domain table had wrong counts in 10 of 16 rows
- Added 3 network_scan_targets to seed_demo.sql for demo completeness
- Added curl examples to Agent Groups section in features.md
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Every major section now explains why the feature matters (not just what it
does) with concrete curl examples. Fixes stale counts: 84→91 endpoints,
18→19 tables, 860→900+ tests, 85→93 OpenAPI operations. Adds network scan
env vars to config reference and M21/M22 rows to feature matrix.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
M21 adds server-side active TLS scanning of CIDR ranges with concurrent
probing, sentinel agent pattern for pipeline reuse, and full CRUD API for
scan targets. M22 adds Prometheus exposition format endpoint alongside
existing JSON metrics. Comprehensive documentation audit updates all docs
to reflect 91 endpoints, 19 tables, 6 scheduler loops, and 900+ tests.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add SOC 2 Type II, PCI-DSS 4.0, and NIST SP 800-57 compliance mapping
guides — the final V2 deliverable. All claims verified against actual
codebase (router.go, config.go, main.go). Also audit and update all
existing docs: fix endpoint/tool/test counts in features.md, expand
demo-guide.md and demo-advanced.md with CLI/MCP/discovery coverage,
update connectors.md F5/IIS status to V3 paid, add compliance reference
to architecture.md.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
shankar0123