Publishing a ranked list of a live system's own weak points - including
explicitly naming the "single highest-value target" - is a real security
problem for a public repo, not a portfolio concern. Removed the section
and the matching stat-table row. Left the fixed cert-renewal incident in
place since there's nothing exploitable in a resolved issue.
Access model now covers the narrowly-scoped read-only utility policies
(ansible sync roles, metrics-reader, pki-reader) alongside the
host-scoped svc-* pattern - the fuller picture confirmed live against
Vault's actual ACL policy list, not just the original narrower claim.
Added two real screenshots: the live policy list, and an actual audit
log entry showing a KV read with every secret value as hmac-sha256:...
instead of plaintext - proof for the audit-trail claim instead of just
an assertion of it.
Removed the before/after plaintext-secrets Mermaid diagram (the bullets
right below it already said the same thing more clearly) and the
Architecture flowchart (its per-host labels were silently missing
registry-token and Hawser-secret details that didn't fit in the boxes).
Replaced the flowchart with an accurate, complete "who gets what" table
sourced directly from Trilium, one sentence for the proxy-routing fact
it was trying to show, and a new sequence diagram for the actual secret
lifecycle (auth -> render -> read -> async rotation) - a genuinely
different kind of diagram since that's process/time-ordered information
a table can't represent well.
Also restored real specificity Trilium had that the doc had flattened
into vague language: the cert-renewal row now names Technitium
specifically and states the actual incident (an expired cert silently
served for 19 days before the auto-rebuild hook existed), and the audit
trail row now states the real rotation policy (weekly, 8 generations).
Removed the dependency-graph Mermaid diagram (had backwards Vault
dependency arrows, contradicted the documented ingress model) and the
four-boundary Mermaid diagram (fully redundant with the table below it).
Verified every remaining claim against Trilium's Traffic Flow Model and
fixed two that didn't hold up: the DNS-bypass failure behavior overstated
an unconditional SSO bypass, and the notifications failure behavior had
an unsourced claim about backlog retry.
Added the real story behind the trust-boundary hardening (three stages
of the same "trusted by default" flaw getting found and closed, not a
changelog of dated facts), a new trust-depth diagram showing that being
"inside the network" still enforces its own internal boundaries, and
real pfSense screenshots proving the LAN lockdown rules actually exist
in production, not just in the doc.
Fixed one real inaccuracy (Gitea registry scope, Secret Distribution
path count) and replaced generic descriptions with true, specific
detail across all 14 Systems bullets - real incidents, real design
decisions, concrete numbers - each checked against Trilium and, where
relevant, the live Ansible roles rather than trusted at face value.
Also stages a verified-but-unintegrated PAM hardening finding in
docs/firewall-and-network.md for that file's own upcoming audit pass.
Removed by request - it was the one unfinished system in the repo,
and it's the user's call on what represents them publicly. Removed
completely: the README section, docs/content-pipeline.md, and the
two stacks that only existed to back it (postiz, nocodb-content).
Fixed cross-links in dns-naming.md and firewall-and-network.md that
pointed at it (replaced the public-facing-app example with the
still-present Ghost blog), and notifications.md (dropped the
reference, also fixed a leftover 'documented rather than hidden'
phrase in the same paragraph). Stat strip and stack counts updated
(55->53 stacks, 16->15 systems).
README is short/sharp facts, docs/*.md is full depth/process/reasoning
- that split had collapsed into either near-duplicate text in both
files or the fact missing from the README entirely. README now states
the OIDC zone-crossing fact in one crisp sentence; the full doc has
the complete step-by-step breakdown. The LDAP superuser workaround
claim is removed from both - the source material frames it as
contingency guidance for a scenario that may never have come up, not
documented deployed behavior, so it doesn't belong in either file.
The source material says header-based forward-auth is 'the viable
path instead, if SSO is wanted for an app like this' - conditional
contingency guidance for a scenario that may not have actually come
up, not a description of something built and running. Presenting it
as live, deployed behavior in both docs/sso-and-identity.md and the
README was a real accuracy error. Removed from both; the doc still
covers what's actually confirmed (Authentik, dual-zone registration,
native OIDC vs. proxy forward-auth split, least-privilege Git access).
Actually fixed the 4 problems named earlier instead of just naming
them: a numbers-at-a-glance strip up top, the trust-boundary diagram
is now color-coded by boundary type, added a concrete blast-radius
table showing exactly what stays up vs goes down per failure, and
replaced vague figures ('~40 stacks', 'every managed host') with
real ones (41 stacks, 11 fleet hosts).
Leads with the hardest actual decision (4 trust boundaries, each
calibrated to a different risk level, enforced by 3 separate routing
authorities) given full narrative depth instead of flat template
treatment. The other 6 traffic laws move to a compact reference
table linking to their own dedicated docs, avoiding redundancy with
content that already lives elsewhere in more depth.
Previously only delivered 2-3 of the 8 traffic categories it claimed
to cover. Now states every law (ingress, east-west, secrets, DNS,
observability, notifications, CI/CD, management) with its own
Law/Flow/Failure-behavior, plus the 4 global architectural laws that
apply across all of them.
Switch to LR layout, restore all 12 observability-reporting edges
(cut them by mistake, they belong in a dependency graph), drop the
repeated 'reports to' label text that was breaking legibility and
let line style (solid=dependency, dotted=async reporting) carry that
distinction instead. Also shortened edge labels on the trust-boundary
diagram to avoid the same clipping issue.
Completes the full depth pass across all 16 docs - real tech names,
why-this-vs-the-alternative reasoning, heavy cross-linking between
every doc, ranked/specific tradeoffs instead of generic hedges.
Same treatment as the security batch - real tech names, 'why this vs
the alternative' reasoning, heavy cross-linking, ranked/specific
tradeoffs instead of generic hedges.
Names real technology throughout (Vault, Authentik, pfSense,
firewalld, Wazuh) instead of staying generic, adds a 'why this
approach vs the alternative' section to every doc, cross-links
heavily between docs instead of treating each system in isolation.
Names real technology (Technitium) instead of staying vague, adds
explicit cross-links to every other system that actually depends on
DNS, explains why self-hosted DNS was chosen, and adds a 'what I'd
improve' section. Template for redoing the rest.
Completes the docs/ set - all 16 systems now have a full deep-dive
page in the diagram+callout+table style. Architecture overview covers
the 4 trust boundaries and 2 traffic planes; IaC pipeline is a fresh
Packer->Terraform->Ansible writeup replacing the stale old-content/
snapshot; disaster recovery covers the 3 independent backup layers
and the deliberately decoupled OS/app upgrade systems.
Completes the orchestration/registry/content batch - Kestra's
check-pause-approve-apply pattern, the self-hosted build pipeline,
the SOC dashboard's 14-source aggregation, the cron heatmap, and the
in-progress content pipeline, all in the diagram+callout+table style.
Mermaid diagrams up front, GFM alert callouts for key decisions and
gotchas, tables instead of bullet walls - replaces plain prose with
the visual template established for security-monitoring.md.
Secrets & PKI, Secret Distribution Model, SSO & Identity,
Firewall & Network Segmentation, DNS Naming Model, and Security
Monitoring - the first batch of per-system write-ups linked from
the new README, sourced and sanitized from the internal KB.
thesabear