Merge the GitHub line: one history, so a tag move survives the sync

hanzoai/ci had two diverged mains — five commits only on github.com (the
vendormark and site-zip-publish lines) and seven only on git.hanzo.ai (the
delegate gate, the publishable rule, the registry-credential fix). Their
merge-base was f098b39, which is exactly where v1 and v2 still pointed, so the
drift was invisible: both tags named the last commit the two lines agreed on.

That is what reverted the first attempt at this. Moving v1/v2 on the forge alone
held for three minutes and was then pushed back to f098b39 by the sync that
mirrors github.com into the forge — runs 101 and 102 are the tags reappearing at
the old commit. A tag is a ref like any other; if one side of a mirrored pair
does not have the commit, the sync restores what it does have. So the histories
have to converge before the tag can move anywhere.

The single conflict was two correct changes to the same lines, and both are
kept. The registry credential is fetched UNCONDITIONALLY (this line): it is not
an object-store credential, and a repo declaring only `images:` was reaching
docker login with two empty strings. The object-store gate narrows to `bucket:`
only (the github line): the site lane no longer writes to S3 — it hands the zip
to cloud — so it no longer needs a key that is not in KMS for any org. Taking
either side alone would have re-broken the other.

Under `mode: buildx` all 24 steps still have effective conditions identical to
f098b39. All ten gates pass, including build-yml-is-one-file, which is the gate
that caught 8a415e9 patching only the .github copy.

.github/actions/sitedeploy/action.yml

@@ -0,0 +1,60 @@
+name: sitedeploy
+description: Publish a built static export to the Hanzo PaaS Sites plane.
+# For every repo whose deploy is "a static export goes live" — hanzo.ai, hanzo.app,
+# hips, computer and the ones after them. Those were each about to carry their own
+# copy of enqueue → upload → complete, which is how one contract becomes N
+# transcriptions that drift (see the imgver action next door for the same story
+# told about image tags).
+#
+#   - uses: hanzoai/ci/.github/actions/sitedeploy@v1
+#     with:  { slug: hanzo-ai, dir: out }
+#     env:   { HANZO_DEPLOY_TOKEN: '${{ secrets.HANZO_DEPLOY_TOKEN }}' }
+#
+# A SITE IS NOT AN APP. This publishes files and stops: no image, no CR, no
+# replicas, no registry. Building a container so a Go binary can serve /public is
+# the shape the Sites plane exists to retire.
+#
+# ONE credential. The 202 hands back a prefix-scoped, 30-minute presigned POST
+# grant, so CI never holds a bucket key — do NOT add SITES_S3_* here; that is the
+# standing shared-bucket credential the grant replaced.
+inputs:
+  slug:
+    description: The project slug on the Sites plane (POST /v1/projects/<slug>/deploy)
+    required: true
+  dir:
+    description: The built export directory
+    required: true
+  api:
+    description: Cloud API base
+    required: false
+    default: https://api.hanzo.ai
+  jobs:
+    description: Parallel uploads
+    required: false
+    default: '24'
+runs:
+  using: composite
+  steps:
+    - name: Fetch sitedeploy
+      shell: bash
+      # The action ref is the script ref: an action pinned to @v1 runs v1's
+      # sitedeploy. Both forges, because this repo is served from each.
+      run: |
+        set -euo pipefail
+        ref="${GITHUB_ACTION_REF:-v1}"
+        for url in https://git.hanzo.ai/hanzoai/ci https://github.com/hanzoai/ci; do
+          git clone -q --depth 1 --branch "$ref" "$url" "$RUNNER_TEMP/sitedeploy-ci" 2>/dev/null && break
+        done
+        [ -x "$RUNNER_TEMP/sitedeploy-ci/bin/sitedeploy" ] \
+          || { echo "::error::could not fetch hanzoai/ci@$ref (bin/sitedeploy)"; exit 1; }
+
+    - name: Deploy
+      shell: bash
+      env:
+        HANZO_API: ${{ inputs.api }}
+        SITEDEPLOY_JOBS: ${{ inputs.jobs }}
+        SITEDEPLOY_COMMIT: ${{ github.sha }}
+        SITEDEPLOY_BRANCH: ${{ github.ref_name }}
+      run: |
+        set -euo pipefail
+        bash "$RUNNER_TEMP/sitedeploy-ci/bin/sitedeploy" '${{ inputs.slug }}' '${{ inputs.dir }}'

.hanzo/workflows/cicd.yml

@@ -23,5 +23,5 @@ concurrency:
   cancel-in-progress: true
 jobs:
   gate:
-    uses: hanzoai/ci/.hanzo/workflows/build.yml@v2
+    uses: hanzoai/ci/.hanzo/workflows/build.yml@v1
     secrets: inherit

LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 Hanzo AI, Inc.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

README.md

@@ -171,10 +171,50 @@ Builds are `CGO_ENABLED=0 -trimpath`: the host that installs this runs it on
 whatever base image the host is, and the digest must be a function of the
 source, not of the checkout path.
 
-## Runners — our cloud or your own
+## `site:` — a static export, promoted to an immutable release
 
-By default the build runs on the **Hanzo cloud** arc pool (we run it; metered as
-build minutes). To run on **your own** self-hosted arc runners, pass their labels:
+`images:` ships an OCI image a **cluster** runs; `binaries:` ships an executable
+a **host** installs. `site:` ships a static export an **edge** serves — no image,
+no CR, no replicas. Building a container so a Go binary can serve `/public` is
+the shape this retires.
+
+```yaml
+site:
+  slug: hanzo-console         # the project on the Sites plane
+  dir: out                    # the built export; needs index.html at its root
+  build: npm ci && npm run build   # optional; run first
+  on: [main]                  # same branch gate as deploy.on; tags always publish
+```
+
+That is the whole configuration. **There is no credential to provision**: the
+bearer is the IAM JWT the workflow already mints from `KMS_CLIENT_ID` /
+`KMS_CLIENT_SECRET`, so a repo that can build can publish. CI names no bucket and
+no org — the org segment is prepended server-side from the validated principal,
+which is what makes the prefix unforgeable.
+
+The export is zipped and posted to `/v1/projects/<slug>/deploy`, then that prefix
+is promoted by `/v1/sites/<slug>/publish` into an **immutable release** whose id
+digests its object manifest. The site's pointer is flipped to it, and the step
+then re-reads the release list and refuses unless the release it just published
+is the one that is live. Rollback is the same pointer aimed at an older release.
+
+**One size boundary, and it is the server's.** cloud's public edge caps a request
+body at 16 MiB (`GATEWAY_BODY_LIMIT`) and refuses a larger POST before any
+handler runs — reporting only `Error when parsing request`, which names neither
+size nor cause. `bin/sitepublish` therefore measures the zip and refuses *with
+the number* first. Of the 24 built exports in the estate, 22 fit; the two that do
+not (`hanzo.ai` at 27.9 MiB zipped and 8,536 files, `trillerfest.com` at 76.7
+MiB) use [`bin/sitedeploy`](bin/sitedeploy), which streams per-file against a
+presigned grant and has no body limit. `hanzo.ai` is also past the server's own
+5,000-entry cap, so no transport makes that export a release.
+
+## Runners — our fleet or your own
+
+By default the build runs on the **Hanzo `git-runner` fleet** on git.hanzo.ai
+(we run it; metered as build minutes) — the only pool that serves the default
+`hanzo-build-linux-amd64` label. There is no arc pool: arc (arcd) was retired
+2026-08-01 and never served any label in this default. To run on **your own**
+self-hosted runners, pass their labels:
 
 ```yaml
     uses: hanzoai/ci/.github/workflows/build.yml@v1
@@ -185,7 +225,7 @@ build minutes). To run on **your own** self-hosted arc runners, pass their label
 
 ## Delegate to platform (skip runner buildx)
 
-By default the build runs buildx **on** the arc runner. To instead hand the build
+By default the build runs buildx **on** the runner. To instead hand the build
 to **platform.hanzo.ai** — which builds in-cluster with BuildKit and rolls the
 service itself — pass `mode: delegate`:
 
@@ -197,7 +237,7 @@ service itself — pass `mode: delegate`:
 ```
 
 The GitHub job then just POSTs each image in `hanzo.yml` to platform's direct
-build webhook (`/v1/arcd/enqueue`) and exits in **seconds** — no runner buildx,
+build webhook (`/v1/runner`) and exits in **seconds** — no runner buildx,
 no KMS, no runner-side deploy. Platform creates the build job, launches an
 in-cluster BuildKit Job on its own pool, pushes to the registry, and patches the
 operator `Service` CR to roll it. It's the same build path as the platform
@@ -205,10 +245,10 @@ GitHub-App webhook — one build path, two front doors.
 
 Requires one extra secret, `PLATFORM_BUILD_CALLBACK_TOKEN` (org- or repo-level,
 picked up via `secrets: inherit`). Override the endpoint with the
-`PLATFORM_ENQUEUE_URL` repo/org variable (default `https://platform.hanzo.ai/v1/arcd/enqueue`).
+`PLATFORM_ENQUEUE_URL` repo/org variable (default `https://platform.hanzo.ai/v1/runner`).
 
 `mode: buildx` (the default) is unchanged — existing repos keep running buildx on
-arc, so delegation is strictly opt-in.
+the fleet runner, so delegation is strictly opt-in.
 
 ## Credentials
 
@@ -220,6 +260,6 @@ run time. No long-lived registry or cluster credentials live in GitHub.
 ## Platform-native
 
 `hanzo.yml` is also read by platform.hanzo.ai: a repo on the platform webhook
-needs **only** `hanzo.yml` — the platform builds it on arc and rolls it out, no
+needs **only** `hanzo.yml` — the platform builds it in-cluster and rolls it out, no
 workflow file at all. This reusable is the GitHub-Actions path for repos that
 trigger through GitHub instead of the platform.

bin/conflictmarkers

@@ -0,0 +1,112 @@
+#!/usr/bin/env bash
+# conflictmarkers — refuse a repo that has committed an unresolved merge.
+# One implementation, every caller.
+#
+#   conflictmarkers [dir]
+#
+# WHAT THIS CATCHES
+#
+# `git merge` writes its disagreement INTO the file and stops. Resolving means
+# editing those lines out; nothing forces you to. `git add` on a file that still
+# contains them is accepted without complaint, and from that moment the markers
+# are ordinary committed content — `git status` is clean, `git diff` is empty,
+# and the conflict is now a permanent feature of the branch.
+#
+# It survives because of WHERE it lands. The file is usually generated or
+# vendored — large, rarely opened, excluded from review by its own size. Nobody
+# reads line 23,807 of a .d.ts. hanzoai/base carried exactly this on main:
+#
+#   plugins/jsvm/internal/types/generated/types.d.ts:23807   <<<<<<< HEAD
+#   plugins/jsvm/internal/types/generated/types.d.ts:23850   >>>>>>> upstream/master
+#
+# from an upstream merge nobody finished, sitting in the shipped package.
+#
+# The compilers are no help, which is the whole problem. In TypeScript the
+# markers are a syntax error — but a .d.ts nothing imports is never parsed, so
+# there is no error to see. In Go they are a parse error only in a file the
+# build reaches. In YAML, JSON, Markdown, SQL, HTML and every config format we
+# ship, they are silently VALID content: a marker in a values file is a key
+# nobody notices, and a marker in a Markdown doc renders as text.
+#
+# So the defect class is "both sides of a merge shipped, and nothing in the
+# pipeline had an opinion". That is the same shape as its two siblings in this
+# step — a declaration and reality disagreeing, with no reader positioned to
+# notice — which is why it belongs here rather than in any one repo's gate.
+#
+# WHAT IT LOOKS FOR, AND WHY NOT MORE
+#
+# Only the two LABELLED markers git actually writes:
+#
+#   ^<<<<<<< <label>     the start marker, seven '<' then a space
+#   ^=======              the divider, seven bare equals
+#   ^>>>>>>> <label>     the end marker, seven '>' then a space
+#
+# ALL THREE, IN THAT ORDER, EACH ON ITS OWN LINE. Not any one of them — the
+# whole shape. That is not belt-and-braces, it is the difference between a gate
+# and a nuisance, and it was measured the expensive way: matching the start
+# marker ALONE failed hanzoai/app, whose builder tests carry
+#
+#   <<<<<<< START_TITLE index.html >>>>>>> END_TITLE
+#
+# as literal fixture data. That line opens with seven '<' and a space and is
+# not a merge conflict — both markers sit on ONE line, which is a thing git
+# never writes. A start marker that also carries an end marker is therefore
+# excluded outright, and a file must show the divider and a separate end line
+# before this refuses it.
+#
+# The divider is only ever matched as part of that conjunction, never alone:
+# seven bare equals signs at column 1 is also how reStructuredText and Setext
+# Markdown underline a heading, and failing every doc in the estate is how a
+# gate gets switched off.
+#
+# Exactly seven, anchored at column 1: `<<<<<<<<` (eight — a heredoc, a C++
+# stream) does not match, and neither does an indented mention inside a comment
+# explaining conflict markers, which is what lets this file describe them
+# without flagging itself.
+#
+# Tracked files only, via `git grep`, so a stray marker in an untracked scratch
+# file or in node_modules cannot fail a build. `-I` skips binaries.
+set -euo pipefail
+
+cd "${1:-.}"
+
+git rev-parse --is-inside-work-tree >/dev/null 2>&1 || {
+  echo "conflictmarkers: not a git work tree ($(pwd)) — skipping"; exit 0; }
+
+# Built from variables rather than written literally, so this script is not its
+# own first finding when it scans the repo that carries it.
+L=$(printf '<%.0s' $(seq 7))
+E=$(printf '=%.0s' $(seq 7))
+R=$(printf '>%.0s' $(seq 7))
+
+# Candidate files: those carrying a start marker that is NOT also an end marker
+# on the same line. `-l` for names only; the line numbers come later, per file.
+cands=$(git grep -I -l -E "^${L} " -- . || true)
+
+bad=""
+for f in $cands; do
+  grep -qE "^${L} " -- "$f" 2>/dev/null || continue
+  # The same-line form (app's fixture) is not a conflict. Require at least one
+  # start line that does not also carry an end marker.
+  grep -E "^${L} " -- "$f" 2>/dev/null | grep -qv -- "$R" || continue
+  # And require the rest of the shape: a bare divider and a separate end line.
+  grep -qE "^${E}$" -- "$f" 2>/dev/null || continue
+  grep -E "^${R} " -- "$f" 2>/dev/null | grep -qv -- "$L" || continue
+  bad="$bad $f"
+done
+
+if [ -n "$bad" ]; then
+  echo "::error::committed merge-conflict markers — an unresolved merge is in the tree"
+  for f in $bad; do
+    grep -nE "^(${L} |${E}$|${R} )" -- "$f" | while IFS= read -r line; do
+      echo "  $f:$line"
+    done
+    echo "::error file=${f}::committed conflict marker"
+  done
+  echo
+  echo "Resolve the merge and commit the result. If the file is GENERATED, do not"
+  echo "hand-edit it — re-run its generator and commit that."
+  exit 1
+fi
+
+echo "OK: no committed conflict markers ($(git ls-files | wc -l) tracked files)"

bin/ignoretracked

@@ -0,0 +1,118 @@
+#!/usr/bin/env bash
+# ignoretracked — refuse a repo that ships bulk content its own .gitignore
+# claims to be ignoring. One implementation, every caller.
+#
+#   ignoretracked [dir]
+#
+# WHAT THIS CATCHES
+#
+# An ignore rule does not untrack what is already tracked. `.gitignore` is
+# consulted when git decides whether to ADD an untracked path; a path already
+# in the index is never reconsidered. So this arrangement is stable and silent:
+#
+#   .gitignore says   native/flags/target/
+#   the index says    855 files under native/flags/target/
+#   git status says   nothing
+#
+# hanzoai/cloud lived there for 6 days and 16 hours. 362 MB of orphaned cargo
+# output — three near-identical 37.8 MB staticlibs and a pile of .rlib — went
+# into every release, took the module past Go's 500 MiB ceiling (see
+# bin/modsize) and made it UNFETCHABLE for nine consecutive releases. The
+# `git archive` of that commit is 202 MB against 12.5 MB today: 16x.
+#
+# The causal story is worth knowing because it is not carelessness, it is a
+# race nobody could see. The ignore rule was REMOVED at 12:22 (the Rust
+# staticlib was being dropped for a Go evaluator), the 855 files were added at
+# 12:51 inside that 4.5-hour window when nothing was ignoring them, and the
+# rule was RE-ADDED at 16:48. Every individual step was reasonable. The state
+# they combined into is the defect, and no tool in the pipeline had an opinion
+# about it.
+#
+# WHY THIS GATES ON BYTES AND NOT ON PRESENCE
+#
+# The obvious gate — "any tracked path matched by .gitignore fails" — is
+# correct in principle and unshippable in fact. Measured across all 845 git
+# repos in the three orgs: 155 of them (18%) carry at least one tracked file
+# their .gitignore matches, and hanzoai/cloud's own origin/main is one of them
+# (a bare, unanchored `tools` pattern on .gitignore:62 catches 27 legitimately
+# tracked Go source files under apps/tools/). `CLAUDE.md` alone is ignored-and-
+# tracked in 40 repos and is the ONLY hit in 29 of them. Turning that on as a
+# hard fail breaks a fifth of the estate on day one, and a gate that fails what
+# ought to pass is a gate that gets switched off — after which we are worse off
+# than before it existed.
+#
+# So the gate is on the quantity that actually caused the outage. The byte
+# distribution separates cleanly, which is why this threshold can be a refusal
+# rather than a warning:
+#
+#     worst legitimate repo in the estate   61.5 MB   (hanzo/docs, vendored)
+#     ... next                              48.8 MB, 45.8, 45.2, 43.8, 39.8
+#     repos over 100 MB                     ZERO
+#     the defect this gate exists for      362.0 MB
+#
+# Default ceiling 100 MB: green on all 845 repos today with 62% headroom over
+# the worst honest case, and red on the real defect by 3.6x. Nothing to
+# baseline, no allowlist, no per-repo exemptions — the tail is REPORTED (so the
+# hygiene problem stays visible and shrinkable) and only the bulk is REFUSED.
+#
+# As repos are cleaned up, lower IGNORETRACKED_MAX_MB. It is a ratchet, and the
+# end state is 0 — at which point this becomes the pure presence gate that was
+# right all along. Do not start there.
+#
+# WHY core.excludesFile IS PINNED OFF
+#
+# `--exclude-standard` reads THREE sources: the repo's committed .gitignore
+# files, .git/info/exclude, and the user's global core.excludesFile. The last
+# is per-machine, so the same commit gets different verdicts on a laptop and a
+# runner — on this workstation a global bare `tags` pattern matched 1,271
+# Elixir source files in lux/explorer-v1 that CI would never flag. A gate whose
+# answer depends on whose machine asked is not a gate. Only committed
+# .gitignore is in scope, so the global file is pinned to /dev/null.
+# (.git/info/exclude is per-clone and is the stock all-comments template on a
+# fresh CI checkout; it cannot be overridden by -c, and is left alone.)
+#
+# EXIT: 0 clean or under the ceiling (findings still reported), 1 over it.
+set -uo pipefail
+
+MAX_MB=${IGNORETRACKED_MAX_MB:-100}
+root=${1:-.}
+cd "$root" 2>/dev/null || { echo "ignoretracked: no such directory: $root" >&2; exit 1; }
+git rev-parse --git-dir >/dev/null 2>&1 || exit 0   # not a repo: nothing to check
+
+tmp=$(mktemp -d); trap 'rm -rf "$tmp"' EXIT
+
+# -s gives "<mode> <sha> <stage>\t<path>", which carries the blob id — so the
+# size comes from cat-file and needs no working-tree stat (correct on a bare or
+# sparse checkout, and unaffected by anything a build has written).
+git -c core.excludesFile=/dev/null ls-files -s --cached --ignored --exclude-standard \
+  > "$tmp/ls" 2>/dev/null || exit 0
+[ -s "$tmp/ls" ] || { echo "ignoretracked: OK — nothing tracked that .gitignore ignores"; exit 0; }
+
+files=$(wc -l < "$tmp/ls")
+bytes=$(awk '{print $2}' "$tmp/ls" \
+  | git cat-file --batch-check='%(objectsize)' --buffer 2>/dev/null \
+  | awk '/^[0-9]+$/{s+=$1} END{print s+0}')
+mb=$(awk -v b="${bytes:-0}" 'BEGIN{printf "%.1f", b/1048576}')
+
+# WHICH .gitignore LINE is doing this. Plain `check-ignore` prints nothing for a
+# tracked path (it answers "would git ignore this if it were untracked", and a
+# tracked path short-circuits) — --no-index is what makes it answer for the
+# files we actually have. The output names source:line:pattern, so the report
+# points at the line to edit instead of at 855 paths to read.
+cut -f2- "$tmp/ls" | git -c core.excludesFile=/dev/null check-ignore --no-index -v --stdin \
+  > "$tmp/why" 2>/dev/null || true
+
+echo "ignoretracked: ${files} tracked file(s), ${mb} MB, matched by this repo's own .gitignore"
+if [ -s "$tmp/why" ]; then
+  echo "  by rule (top 10):"
+  awk -F'\t' '{n[$1]++} END{for (k in n) printf "%8d  %s\n", n[k], k}' "$tmp/why" \
+    | sort -rn | head -10
+fi
+
+limit=$(( MAX_MB * 1048576 ))
+if [ "${bytes:-0}" -ge "$limit" ]; then
+  echo "::error::${mb} MB of tracked content is matched by this repo's own .gitignore — over the ${MAX_MB} MB ceiling. An ignore rule does not untrack what is already tracked, so this ships in every release and every module zip while \`git status\` stays clean. Fix: \`git rm -r --cached <path>\` and commit (the working tree is untouched). If the content is meant to be tracked, un-ignore it instead — negate the rule with \`!<path>\` — so the two stop disagreeing."
+  exit 1
+fi
+echo "ignoretracked: OK — ${mb} MB is under the ${MAX_MB} MB ceiling (reported, not blocking)"
+exit 0

bin/modsize

@@ -0,0 +1,123 @@
+#!/usr/bin/env bash
+# modsize — refuse a Go module that is approaching the size at which Go can no
+# longer fetch it. One implementation, every caller.
+#
+#   modsize [dir]        # default: every module in the repo, from the root
+#
+# WHAT THIS CATCHES
+#
+# Go's module ceiling is not a soft limit and not a warning. From
+# golang.org/x/mod/zip (and, identically, cmd/go/internal/modfetch/codehost):
+#
+#   MaxZipFile = 500 << 20     // 524288000 bytes, 500 MiB
+#
+# and it is enforced THREE ways on the same number — the zip file itself, the
+# total uncompressed size of the files inside it, and (in CheckDir) the running
+# sum of the source tree's file sizes:
+#
+#   "module source tree too large (max size is 524288000 bytes)"
+#   "total uncompressed size of module contents too large (...)"
+#
+# hanzoai/cloud crossed it and became UNFETCHABLE for nine consecutive
+# releases. Nothing said so. The module publishes fine — the ceiling is
+# enforced on the CONSUMER, at `go get`, so the failure surfaces in someone
+# else's repo, at a version they did not choose, long after the release that
+# caused it. There is no signal at the publishing end at all, which is why nine
+# releases went out before anybody knew.
+#
+# WHY THIS SUMS RAW BYTES AND WHY THAT IS EXACT, NOT AN APPROXIMATION
+#
+# It would be reasonable to assume the limit is on the compressed zip and that
+# summing uncompressed bytes over-counts. It does not: zip.CheckDir accumulates
+# `info.Size()` — the raw, uncompressed size of each regular file — and fails
+# when that running total exceeds MaxZipFile. So the sum of file sizes IS one
+# of the three quantities Go bounds, and a git blob's size is exactly that
+# file's uncompressed size. This gate therefore measures the same number Go
+# measures, not a proxy for it.
+#
+# WHAT IT EXCLUDES, AND WHY EACH EXCLUSION IS REQUIRED FOR CORRECTNESS
+#
+#   • Nested modules. A subdirectory with its own go.mod is a DIFFERENT module
+#     and its bytes are not in the parent's zip. Counting them would fail a
+#     repo that is nowhere near the limit — hanzoai/s3 is this shape (it builds
+#     s3-rdma-sidecar/ and telemetry/server/ from their own go.mod files), and
+#     a gate that fails a repo that would have worked is a gate people learn to
+#     skip. Each nested module is instead checked on its own terms.
+#   • Symlinks and gitlinks (submodules). zip.CheckDir skips both — symlinks
+#     explicitly (golang.org/issue/27093), submodules because they are not
+#     files. Counting a gitlink's 20-byte entry would be harmless; counting a
+#     symlink's target as content would not.
+#
+# It reads tracked content only (git ls-tree), because that is what a module
+# zip is built from: the proxy serves what the VCS has at that tag, not what a
+# working tree happens to contain.
+#
+# THE THRESHOLD IS BELOW THE CEILING ON PURPOSE
+#
+# Failing AT 500 MiB would be useless — at that point the module is already
+# unfetchable and the only question left is how many releases shipped broken.
+# The gate refuses at MODSIZE_MAX_PCT (default 80%, ~419 MiB), which is the
+# "warn well before" — spelled as a refusal, because a warning about a cliff
+# nobody is watching is the same defect one level up. 105 MiB of headroom is
+# several releases' worth of honest growth, so this fires with room to fix it
+# calmly and never fires the release it would have broken.
+#
+# EXIT: 0 clean, 1 when a module is at or past the threshold.
+set -uo pipefail
+
+# 500 << 20, from golang.org/x/mod/zip. Not a guess and not rounded: the
+# constant is quoted so a reader can check it against the source.
+CEILING=$((500 << 20))
+PCT=${MODSIZE_MAX_PCT:-80}
+REF=${MODSIZE_REF:-HEAD}
+
+root=${1:-.}
+cd "$root" 2>/dev/null || { echo "modsize: no such directory: $root" >&2; exit 1; }
+git rev-parse --git-dir >/dev/null 2>&1 || exit 0   # not a repo: nothing to measure
+
+# Every go.mod in the tree, at REF. The root module is "go.mod"; anything else
+# is a nested module and marks a prefix the parent must not count.
+mods=$(git ls-tree -r --name-only "$REF" 2>/dev/null | grep -E '(^|/)go\.mod$' | sort) || exit 0
+[ -z "$mods" ] && exit 0
+
+# path<TAB>size for every regular tracked file. Mode 100644/100755 only:
+# 120000 is a symlink and 160000 a gitlink, both of which zip.CheckDir omits.
+# `git ls-tree -l` separates the path with a TAB, so split on that rather than
+# on whitespace: a path with spaces in it must survive intact.
+sizes=$(git ls-tree -r -l "$REF" 2>/dev/null \
+  | sed -n 's/^\([0-9]\{6\}\) blob \([0-9a-f]*\) *\([0-9-]*\)\t\(.*\)$/\1\t\3\t\4/p' \
+  | awk -F'\t' '$1=="100644"||$1=="100755"{print $3 "\t" $2}')
+
+human() { awk -v b="$1" 'BEGIN{ printf "%.1f MiB", b/1048576 }'; }
+
+rc=0
+for gm in $mods; do
+  if [ "$gm" = "go.mod" ]; then mdir=""; else mdir="${gm%/go.mod}/"; fi
+  # The prefixes this module must NOT count: every OTHER module nested under it.
+  nested=$(printf '%s\n' "$mods" | while read -r o; do
+    [ "$o" = "$gm" ] && continue
+    od="${o%go.mod}"
+    case "$od" in "$mdir"?*) printf '%s\n' "$od";; esac
+  done)
+  total=$(printf '%s\n' "$sizes" | awk -F'\t' -v m="$mdir" -v nl="$nested" '
+    BEGIN{ n=split(nl, arr, "\n") }
+    {
+      p=$1
+      if (m != "" && index(p, m) != 1) next          # not in this module
+      for (i=1; i<=n; i++) if (arr[i] != "" && index(p, arr[i]) == 1) next  # nested module
+      s += $2
+    }
+    END{ print s+0 }')
+  limit=$(( CEILING * PCT / 100 ))
+  name=${mdir:-./}
+  if [ "$total" -ge "$CEILING" ]; then
+    echo "::error file=${gm}::module ${name} is $(human "$total") of tracked content — PAST Go's $(human $CEILING) module ceiling. It is already unfetchable: \`go get\` fails with 'module source tree too large (max size is ${CEILING} bytes)' in every consumer, at every version that carries it. Find the bulk with: git ls-tree -r -l ${REF} | sort -k4 -n | tail -20"
+    rc=1
+  elif [ "$total" -ge "$limit" ]; then
+    echo "::error file=${gm}::module ${name} is $(human "$total") of tracked content — ${PCT}% of Go's $(human $CEILING) ceiling, $(human $((CEILING-total))) of headroom left. Past the ceiling the module stops being fetchable AT THE CONSUMER, so the break shows up in someone else's build and not in this one. Find the bulk with: git ls-tree -r -l ${REF} | sort -k4 -n | tail -20"
+    rc=1
+  else
+    echo "modsize: OK — ${name} $(human "$total") ($(( total * 100 / CEILING ))% of $(human $CEILING))"
+  fi
+done
+exit $rc

bin/publishable

@@ -0,0 +1,46 @@
+#!/usr/bin/env bash
+# publishable — refuse a `build_secrets` name that does not declare itself public.
+#
+# Usage: publishable <hanzo.yml>   (reads images[].build_secrets, rc=1 on refusal)
+#
+# WHY A NAME AND NOT A VALUE. A build_secret is handed to buildx as
+# `--build-arg NAME=value`, and a build-arg is IN THE PUBLISHED IMAGE: `docker
+# history` prints it to anyone who can pull. So the only value that may sit
+# here is one that is public on purpose — a Vite/Next static export has no
+# server to read an env from, so its ingest key must be inlined at build, and
+# inlining it is what publishing it means.
+#
+# The value cannot be the test. At the moment ci reads hanzo.yml the value does
+# not exist yet (KMS has not been called), and guessing secrecy from a string's
+# shape is a heuristic that is wrong in both directions. The name is what a
+# reviewer reads, it is in git, and it is decided by the person who knows the
+# answer. So the name carries the assertion.
+#
+# The fleet had already started saying it this way — hanzoai/docs renamed
+# EVENT_INGEST_KEY -> PUBLISHABLE_KEY, and hanzoai/world's four are VITE_*,
+# which a bundler inlines by construction. This turns that convention into the
+# mechanism. Before it, the ONE repo whose value is genuinely publishable
+# (hanzoai/ui, a `pk-…` key) asserted the prefix inside its own Dockerfile, so
+# the check existed once, for one image, and any repo adding a real credential
+# got no check at all.
+set -uo pipefail
+
+f=${1:-hanzo.yml}
+[ -r "$f" ] || exit 0   # no config, nothing declared, nothing to refuse
+
+names=$(yq -r '[(.images // [])[] | (.build_secrets // [])[]] | unique | .[]' "$f" 2>/dev/null) || exit 0
+[ -n "$names" ] || exit 0
+
+rc=0
+for n in $names; do
+  case "$n" in
+    # Prefixes a bundler already treats as client-side, plus an explicit
+    # self-declaration for everything else. Anything outside this set has not
+    # claimed to be public, so it is not baked.
+    PUBLISHABLE_*|PUBLIC_*|NEXT_PUBLIC_*|EXPO_PUBLIC_*|NUXT_PUBLIC_*|VITE_*|REACT_APP_*|*_PUBLISHABLE|*_PUBLIC) ;;
+    *)
+      echo "::error::build_secret '$n' does not declare itself publishable, and a build_secret is baked into the image as a --build-arg where \`docker history\` reveals it. Rename it (PUBLISHABLE_*, PUBLIC_*, NEXT_PUBLIC_*, VITE_*, REACT_APP_*) if the value is public on purpose; if it is a real credential it cannot be a build_secret at all." >&2
+      rc=1 ;;
+  esac
+done
+exit $rc

bin/publishable_test.sh

@@ -0,0 +1,65 @@
+#!/usr/bin/env bash
+# Tests for bin/publishable. Offline and deterministic: every case is a
+# hanzo.yml written into a temp dir. Run: bash bin/publishable_test.sh
+set -uo pipefail
+cd "$(dirname "$0")/.."
+PUB="$PWD/bin/publishable"
+tmp=$(mktemp -d); trap 'rm -rf "$tmp"' EXIT
+fail=0
+
+# t <name> <want-rc> <build_secrets yaml-inline list>
+t() {
+  local name=$1 want=$2 list=$3
+  local d="$tmp/$RANDOM$RANDOM"; mkdir -p "$d"
+  { echo 'images:'; echo '  - name: app'; echo '    repo: ghcr.io/hanzoai/app'
+    [ -n "$list" ] && echo "    build_secrets: $list"; } > "$d/hanzo.yml"
+  out=$(bash "$PUB" "$d/hanzo.yml" 2>&1); rc=$?
+  if [ "$rc" = "$want" ]; then printf 'ok    %-56s rc=%s\n' "$name" "$rc"
+  else printf 'FAIL  %-56s rc=%s (want %s)\n      %s\n' "$name" "$rc" "$want" "$out"; fail=1; fi
+}
+
+echo "--- refused: a name that never claimed to be public ---"
+# The live case. hanzoai/ui declares exactly this, and its value IS publishable
+# — but nothing outside its own Dockerfile could know that.
+t "EVENT_INGEST_KEY is refused"              1 '[EVENT_INGEST_KEY]'
+t "a real credential is refused"             1 '[STRIPE_SECRET_KEY]'
+t "a token is refused"                       1 '[GITHUB_TOKEN]'
+t "a password is refused"                    1 '[DB_PASSWORD]'
+t "a private key is refused"                 1 '[SIGNING_PRIVATE_KEY]'
+t "one bad name among good ones is refused"  1 '[VITE_GTM_ID, EVENT_INGEST_KEY]'
+
+echo "--- allowed: the name declares it ---"
+# These are the fleet's real declarations, verbatim.
+t "PUBLISHABLE_KEY (hanzoai/docs)"           0 '[PUBLISHABLE_KEY]'
+t "VITE_MAPBOX_TOKEN (hanzoai/world)"        0 '[VITE_MAPBOX_TOKEN]'
+t "VITE_SENTRY_DSN (hanzoai/world)"          0 '[VITE_SENTRY_DSN]'
+t "world's four together"                    0 '[VITE_MAPBOX_TOKEN, VITE_SENTRY_DSN, VITE_ANALYTICS_WEBSITE_ID, VITE_GTM_ID]'
+t "NEXT_PUBLIC_ prefix"                      0 '[NEXT_PUBLIC_INGEST_KEY]'
+t "REACT_APP_ prefix"                        0 '[REACT_APP_MAP_KEY]'
+t "EXPO_PUBLIC_ prefix"                      0 '[EXPO_PUBLIC_API_KEY]'
+t "NUXT_PUBLIC_ prefix"                      0 '[NUXT_PUBLIC_API_KEY]'
+t "PUBLIC_ prefix"                           0 '[PUBLIC_ANALYTICS_ID]'
+t "_PUBLISHABLE suffix"                      0 '[STRIPE_PUBLISHABLE]'
+t "_PUBLIC suffix"                           0 '[ANALYTICS_ID_PUBLIC]'
+
+echo "--- silent: nothing declared, nothing to say ---"
+# 44 of the fleet's 47 repos are this case and must be byte-for-byte unchanged.
+t "no build_secrets key at all"              0 ''
+t "empty build_secrets list"                 0 '[]'
+
+echo "--- a missing file is not a refusal ---"
+out=$(bash "$PUB" "$tmp/does-not-exist.yml" 2>&1); rc=$?
+if [ "$rc" = 0 ]; then printf 'ok    %-56s rc=0\n' "absent hanzo.yml is silent"
+else printf 'FAIL  %-56s rc=%s\n' "absent hanzo.yml is silent" "$rc"; fail=1; fi
+
+echo "--- the refusal says what to do about it ---"
+d="$tmp/msg"; mkdir -p "$d"
+printf 'images:\n  - name: app\n    build_secrets: [EVENT_INGEST_KEY]\n' > "$d/hanzo.yml"
+out=$(bash "$PUB" "$d/hanzo.yml" 2>&1)
+for pat in "EVENT_INGEST_KEY" "docker history" "PUBLISHABLE_" "cannot be a build_secret"; do
+  if printf '%s' "$out" | grep -qF "$pat"; then printf 'ok    %-56s\n' "message names '$pat'"
+  else printf 'FAIL  %-56s\n      got: %s\n' "message names '$pat'" "$out"; fail=1; fi
+done
+
+[ "$fail" = 0 ] && echo "PASS" || echo "FAIL"
+exit $fail

bin/sitedeploy

@@ -0,0 +1,238 @@
+#!/usr/bin/env bash
+# sitedeploy — publish a built static export to the Hanzo PaaS Sites plane.
+# One implementation, every static site.
+#
+#   sitedeploy <slug> <dir>      # e.g. sitedeploy hanzo-ai out
+#
+# WHY THIS IS A SCRIPT AND NOT INLINE SHELL: the same forty lines of enqueue →
+# upload → complete were about to be pasted into hanzo.ai, hanzo.app, hips,
+# computer and every static surface after them. Written N times it is right N-1
+# times and then wrong once — which is exactly how `sha-<short>` became the only
+# tag eleven repos ever published (see bin/imgver). The plane's contract lives
+# here, once.
+#
+# THE THREE STEPS, and why the bytes never pass through the API:
+#
+#   POST /v1/projects/<slug>/deploy            -> 202 {id, bucket, prefix, upload}
+#   POST <upload.url> per file                 -> the bytes, straight to S3
+#   POST /v1/projects/<slug>/deployments/<id>/complete
+#
+# A real export is large — hanzo.ai is 128 MB across 8403 files — and cloud's
+# BodyLimit is 16 MiB, so an artifact POST is refused by fasthttp BEFORE any
+# handler runs, and it fails as an opaque 400 "Error when parsing request" that
+# reads like a malformed payload rather than a size cap. The git source is the
+# documented route for exactly this shape.
+#
+# NO STANDING S3 CREDENTIAL. The 202 carries `upload`: a presigned POST policy
+# that is prefix-scoped (starts-with $key "<org>/<slug>/", enforced by S3 itself),
+# short-lived (30 min) and size-bounded. That replaced handing every repo the
+# bucket's own long-lived access key — one key for a bucket whose only tenant
+# separation is the key prefix, so every repo holding it could overwrite EVERY
+# org's site (cloud apps/projects/grant.go). Do not reintroduce SITES_S3_* here.
+# HANZO_DEPLOY_TOKEN is the ONE credential this needs.
+#
+# DELETION IS THE SERVER'S. The grant authorizes writes only, so CI cannot remove
+# a file; `keys` in the completion is the manifest cloud reconciles the prefix
+# against, and it deletes what the build no longer produces. That is where
+# `aws s3 sync --delete` went. It fails CLOSED on an empty manifest, and so does
+# this script — a build that enumerated nothing has failed, and honouring it
+# literally would delete the live site.
+#
+# ENV: HANZO_DEPLOY_TOKEN (required)  the org-scoped `sk-` key; mint at POST /v1/keys
+#      HANZO_API           (https://api.hanzo.ai)
+#      SITEDEPLOY_JOBS     (24)  parallel uploads
+#      SITEDEPLOY_COMMIT / SITEDEPLOY_BRANCH  recorded on the deployment
+#      SITEDEPLOY_PLAN=1   print the manifest and exit; no network. The test seam.
+set -euo pipefail
+
+slug="${1:?usage: sitedeploy <slug> <dir>}"
+dir="${2:?usage: sitedeploy <slug> <dir>}"
+api="${HANZO_API:-https://api.hanzo.ai}"
+jobs="${SITEDEPLOY_JOBS:-24}"
+
+[ -d "$dir" ] || { echo "::error::$dir is not a directory — the build produced no export"; exit 1; }
+
+# ---- content type -----------------------------------------------------------
+# The presigned POST carries no Content-Type condition, so whatever CI sends is
+# what the object stores — and what it stores is what the edge serves. Send
+# nothing and every page is application/octet-stream, which a browser DOWNLOADS
+# instead of rendering: a green deploy that serves an unusable site.
+ctype() {
+  case "${1##*.}" in
+    html|htm) echo 'text/html; charset=utf-8' ;;
+    css)      echo 'text/css; charset=utf-8' ;;
+    js|mjs)   echo 'text/javascript; charset=utf-8' ;;
+    json)     echo 'application/json; charset=utf-8' ;;
+    xml)      echo 'application/xml; charset=utf-8' ;;
+    txt)      echo 'text/plain; charset=utf-8' ;;
+    svg)      echo 'image/svg+xml' ;;
+    png)      echo 'image/png' ;;
+    jpg|jpeg) echo 'image/jpeg' ;;
+    gif)      echo 'image/gif' ;;
+    webp)     echo 'image/webp' ;;
+    avif)     echo 'image/avif' ;;
+    ico)      echo 'image/x-icon' ;;
+    woff)     echo 'font/woff' ;;
+    woff2)    echo 'font/woff2' ;;
+    ttf)      echo 'font/ttf' ;;
+    otf)      echo 'font/otf' ;;
+    wasm)     echo 'application/wasm' ;;
+    pdf)      echo 'application/pdf' ;;
+    webmanifest) echo 'application/manifest+json' ;;
+    map)      echo 'application/json' ;;
+    *)        echo 'application/octet-stream' ;;
+  esac
+}
+
+# ---- cache control ----------------------------------------------------------
+# Mirrors cloud's apps/sites.CacheControlFor so a site cached the same whichever
+# route its bytes took. bin/sitedeploy_test.sh pins these strings; if the server
+# changes its policy the test is what catches the drift.
+#
+# A fingerprinted asset (Vite/Next/webpack emit app.4f3a9c21.js) is immutable: a
+# new build changes the hash, so the old URL can be cached forever.
+#
+# The class is written `[._-]` with the dash LAST and the pattern held in a
+# variable. Go's regexp spells the same set `[.\-_]`, and transcribing that
+# literally into `[[ =~ ]]` is a silent defect twice over: a backslash is literal
+# inside a POSIX bracket expression, so `\-_` reads as the RANGE \…_ and the shell
+# rejects it as "invalid character range" — and because the `if` merely evaluates
+# false, every fingerprinted asset quietly fell back to max-age=3600 instead of
+# immutable. An unquoted inline pattern also has its own quoting hazards; the
+# variable form is the one that is read as a regex rather than a glob.
+fingerprint_re='[._-][0-9a-fA-F]{8,}\.[a-z0-9]+$'
+cachectl() {
+  local k="$1" base="${1##*/}"
+  case "${k##*.}" in
+    html|htm) echo 'public, max-age=60, s-maxage=86400' ;;
+    js|mjs|css|woff|woff2|png|jpg|jpeg|gif|svg|webp|avif|ico|ttf|otf|wasm|data|pck|unityweb|mem)
+      if [[ "$base" =~ $fingerprint_re ]]; then
+        echo 'public, max-age=31536000, immutable'
+      else
+        echo 'public, max-age=3600'
+      fi ;;
+    *) echo 'public, max-age=3600' ;;
+  esac
+}
+
+# ---- the manifest -----------------------------------------------------------
+# Paths RELATIVE to the export root, which is exactly what the completion's
+# `keys` must carry (cloud reconciles `keep[rel]` against them).
+#
+# CNAME does not travel. It is a GitHub Pages artifact that means nothing to S3
+# and would ship a stale hostname claim into the bucket.
+manifest=$(cd "$dir" && find . -type f ! -name CNAME | sed 's|^\./||' | LC_ALL=C sort)
+count=$(printf '%s' "$manifest" | grep -c . || true)
+[ "$count" -gt 0 ] || { echo "::error::$dir contains no files — refusing to deploy an empty manifest"; exit 1; }
+
+if [ -n "${SITEDEPLOY_PLAN:-}" ]; then
+  printf 'slug=%s dir=%s files=%s\n' "$slug" "$dir" "$count"
+  while IFS= read -r k; do [ -n "$k" ] && printf '%s\t%s\t%s\n' "$k" "$(ctype "$k")" "$(cachectl "$k")"; done <<< "$manifest"
+  exit 0
+fi
+
+: "${HANZO_DEPLOY_TOKEN:?HANZO_DEPLOY_TOKEN is unset. It is the ONE credential this needs; set it on the forge (git.hanzo.ai), which is what reads .hanzo/workflows — not on GitHub. Mint: POST /v1/keys {\"type\":\"secret\"}}"
+
+# ---- 0. ensure the project exists (self-provision) --------------------------
+# A brand-new slug has no project, and the enqueue below 404s on a missing one —
+# which is why every never-before-deployed site failed its first run. Create it
+# idempotently here instead: 200/201 the first time, 409 every time after, both
+# success. So a new site needs no out-of-band `POST /v1/projects`; committing the
+# workflow is enough. The repo is linked because the git-source enqueue requires
+# a linked project (deployGit rejects an unlinked one); SITEDEPLOY_REPO overrides
+# the origin for a forge checkout whose remote is not the canonical repo.
+repo="${SITEDEPLOY_REPO:-$(git remote get-url origin 2>/dev/null || true)}"
+pcode=$(curl -sS -o /tmp/sd-proj.json -w '%{http_code}' \
+  -X POST "$api/v1/projects" \
+  -H "Authorization: Bearer $HANZO_DEPLOY_TOKEN" -H 'Content-Type: application/json' \
+  -d "$(jq -nc --arg s "$slug" --arg u "$repo" --arg b "${SITEDEPLOY_BRANCH:-main}" \
+        '{slug:$s, name:$s} + (if $u=="" then {} else {repo:{url:$u, branch:$b}} end)')")
+case "$pcode" in
+  200|201|409) : ;;  # created now, or already there — either is the state we need
+  *) echo "::error::ensure project $slug returned HTTP $pcode"; head -c 400 /tmp/sd-proj.json; echo; exit 1 ;;
+esac
+
+# ---- 1. enqueue -------------------------------------------------------------
+# 202 Accepted is the success code: the deployment is queued, not live. bucket
+# and prefix come FROM cloud (sitePrefix(org, slug) is server-side) — never guess
+# them, or the upload lands where nothing is served the moment an org or slug
+# changes.
+code=$(curl -sS -o /tmp/sd-enq.json -w '%{http_code}' \
+  -X POST "$api/v1/projects/$slug/deploy" \
+  -H "Authorization: Bearer $HANZO_DEPLOY_TOKEN" -H 'Content-Type: application/json' \
+  -d "{\"source\":\"git\",\"commit\":\"${SITEDEPLOY_COMMIT:-}\",\"branch\":\"${SITEDEPLOY_BRANCH:-main}\"}")
+if [ "$code" != "202" ]; then
+  echo "::error::enqueue $api/v1/projects/$slug/deploy returned HTTP $code"; head -c 600 /tmp/sd-enq.json; echo; exit 1
+fi
+dep=$(jq -r '.id' /tmp/sd-enq.json)
+prefix=$(jq -r '.prefix' /tmp/sd-enq.json)
+upload_url=$(jq -r '.upload.url // empty' /tmp/sd-enq.json)
+if [ -z "$upload_url" ]; then
+  echo "::error::the 202 carried no upload grant, so there is no way to write the bytes."
+  echo "         cloud mints one only when presigning is configured (S3_ADMIN_* on the cloud deployment)."
+  exit 1
+fi
+jq -c '{id,version,status,bucket,prefix}' /tmp/sd-enq.json
+
+# A build that dies after this point would leave the deployment "queued" and the
+# project stuck "building" forever. Report the failure so cloud records an honest
+# terminal state instead of a lie by omission.
+fail() {
+  curl -sS -X POST "$api/v1/projects/$slug/deployments/$dep/complete" \
+    -H "Authorization: Bearer $HANZO_DEPLOY_TOKEN" -H 'Content-Type: application/json' \
+    -d '{"status":"error","message":"CI upload failed"}' >/dev/null 2>&1 || true
+  echo "::error::marked deployment $dep as error"
+}
+trap 'fail' ERR
+
+# ---- 2. the bytes -----------------------------------------------------------
+# `file` goes LAST: S3 ignores every field after the file part, so a grant field
+# trailing the body is silently dropped and the signature check fails.
+#
+# `key` is DROPPED from the grant's fields and re-sent per object. The grant
+# carries key="<org>/<slug>/" — the starts-with PLACEHOLDER, not a destination —
+# and forwarding it verbatim alongside the real key posts `key` twice, which S3
+# answers 400 for every object. That is the whole of the first end-to-end run:
+# 8403 files, 8403 400s. Everything else in the map (bucket, policy, x-amz-*) is
+# covered by the signature and must travel untouched.
+jq -r '.upload.fields | to_entries[] | select(.key != "key") | "-F\n\(.key)=\(.value)"' /tmp/sd-enq.json > /tmp/sd-fields
+put() {
+  local rel="$1" args=() line
+  # A read loop, NOT `mapfile`: mapfile is bash 4+, and macOS ships bash 3.2, so
+  # on a dev box it fails as `command not found`, the array stays EMPTY, and every
+  # upload goes out with no policy or signature at all. S3 then rejects it for a
+  # malformed X-Amz-Credential — an error that points at the credential rather
+  # than at the array that never got built.
+  while IFS= read -r line; do args+=("$line"); done < /tmp/sd-fields
+  curl -sS --fail-with-body -o /dev/null \
+    -X POST "$UP_URL" \
+    -F "key=$PREFIX/$rel" "${args[@]}" \
+    -F "Content-Type=$(ctype "$rel")" -F "Cache-Control=$(cachectl "$rel")" \
+    -F "file=@$DIR/$rel" \
+    || { echo "::error::upload failed: $rel"; return 1; }
+}
+export -f put ctype cachectl
+export UP_URL="$upload_url" PREFIX="$prefix" DIR="$dir" fingerprint_re
+
+echo "uploading $count files to s3://$(jq -r .bucket /tmp/sd-enq.json)/$prefix ($jobs parallel)"
+printf '%s\n' "$manifest" | grep . | xargs -P "$jobs" -I{} bash -c 'put "$@"' _ {}
+
+# ---- 3. flip it live --------------------------------------------------------
+# `wc -c`, not `stat`: the size flag is spelled -f%z on BSD and -c%s on GNU, and a
+# `stat -f%z || stat -c%s` fallback does not work — on Linux the first arm fails
+# INSIDE the pipeline, awk still exits 0, and the `||` never fires, so the byte
+# count silently reports 0. wc is the one spelling both agree on.
+bytes=$(cd "$dir" && find . -type f ! -name CNAME -exec wc -c {} + | awk '$2!="total"{s+=$1} END{print s+0}')
+jq -n --arg c "${SITEDEPLOY_COMMIT:-}" --argjson f "$count" --argjson b "${bytes:-0}" \
+      --args '{status:"live",commit:$c,files:$f,bytes:$b,keys:$ARGS.positional}' \
+      $(printf '%s\n' "$manifest" | grep .) > /tmp/sd-done.json
+
+code=$(curl -sS -o /tmp/sd-resp.json -w '%{http_code}' \
+  -X POST "$api/v1/projects/$slug/deployments/$dep/complete" \
+  -H "Authorization: Bearer $HANZO_DEPLOY_TOKEN" -H 'Content-Type: application/json' \
+  --data-binary @/tmp/sd-done.json)
+trap - ERR
+if [ "$code" != "200" ]; then
+  echo "::error::complete returned HTTP $code"; head -c 600 /tmp/sd-resp.json; echo; fail; exit 1
+fi
+jq -c '{status,liveUrl,version,files,bytes}' /tmp/sd-resp.json

bin/sitedeploy_test.sh

@@ -0,0 +1,80 @@
+#!/usr/bin/env bash
+# Tests for bin/sitedeploy. Runs OFFLINE: SITEDEPLOY_PLAN=1 stops the script
+# before the first network call and prints the manifest it would upload, so every
+# case here is deterministic and needs no token, no bucket and no cluster.
+# Run: bash bin/sitedeploy_test.sh
+set -uo pipefail
+cd "$(dirname "$0")/.."
+SD="$PWD/bin/sitedeploy"
+tmp=$(mktemp -d); trap 'rm -rf "$tmp"' EXIT
+fail=0
+
+plan() { SITEDEPLOY_PLAN=1 bash "$SD" a-slug "$1" 2>&1; }
+t() { # t <name> <got> <want>
+  if [ "$2" = "$3" ]; then printf 'ok    %-56s -> %s\n' "$1" "$2"
+  else printf 'FAIL  %-56s -> %s (want %s)\n' "$1" "$2" "$3"; fail=1; fi
+}
+# field <dir> <key> <col>  — the ctype (2) or cachectl (3) column for one key
+field() { plan "$1" | awk -F'\t' -v k="$2" -v c="$3" '$1==k{print $c}'; }
+
+site="$tmp/site"; mkdir -p "$site/assets" "$site/nested/deep"
+echo '<h1>hi</h1>'  > "$site/index.html"
+echo 'body{}'       > "$site/assets/app.4f3a9c21.css"
+echo 'x'            > "$site/assets/plain.css"
+echo 'y'            > "$site/assets/chunk-AB12CD34.js"
+echo '{}'           > "$site/data.json"
+echo 'z'            > "$site/nested/deep/page.html"
+echo 'hanzo.ai'     > "$site/CNAME"
+
+# --- the manifest ------------------------------------------------------------
+# Keys are RELATIVE to the export root: cloud reconciles keep[rel] against them,
+# so a leading ./ or an absolute path would match nothing and the completion
+# would prune the entire live site.
+t "keys are relative, no leading ./"  "$(plan "$site" | awk -F'\t' 'NR>1&&$1~/^\.?\//{print "ABS"}' | head -1)" ""
+t "nested paths keep their subdirs"   "$(plan "$site" | awk -F'\t' '$1=="nested/deep/page.html"{print "yes"}')" "yes"
+# CNAME is a GitHub Pages artifact: it means nothing to S3 and would ship a stale
+# hostname claim into the bucket.
+t "CNAME does not travel"             "$(plan "$site" | awk -F'\t' '$1=="CNAME"{print "leaked"}')" ""
+t "file count excludes CNAME"         "$(plan "$site" | head -1 | grep -o 'files=[0-9]*')" "files=6"
+
+# --- content type ------------------------------------------------------------
+# The presigned POST carries no Content-Type condition, so what CI sends is what
+# the object stores and what the edge serves. Send nothing and a browser
+# DOWNLOADS every page instead of rendering it.
+t "html"  "$(field "$site" index.html 2)"              "text/html; charset=utf-8"
+t "css"   "$(field "$site" assets/plain.css 2)"        "text/css; charset=utf-8"
+t "js"    "$(field "$site" assets/chunk-AB12CD34.js 2)" "text/javascript; charset=utf-8"
+t "json"  "$(field "$site" data.json 2)"               "application/json; charset=utf-8"
+
+# --- cache control: mirrors cloud apps/sites.CacheControlFor -----------------
+# These strings are the SERVER's policy, pinned here so the two cannot drift
+# apart silently. If cloud changes CacheControlFor, this is what goes red.
+t "html is short-lived, long at the edge" "$(field "$site" index.html 3)"        "public, max-age=60, s-maxage=86400"
+t "unfingerprinted asset is an hour"      "$(field "$site" assets/plain.css 3)"  "public, max-age=3600"
+# The regression this pins: Go spells the class [.\-_], and transcribing that
+# into [[ =~ ]] makes the shell reject it as an invalid character range. The `if`
+# then merely evaluates false, so every hashed asset silently lost `immutable`.
+t "fingerprinted .hash. is immutable"  "$(field "$site" assets/app.4f3a9c21.css 3)"   "public, max-age=31536000, immutable"
+t "fingerprinted -HASH- is immutable"  "$(field "$site" assets/chunk-AB12CD34.js 3)"  "public, max-age=31536000, immutable"
+
+# --- fail closed -------------------------------------------------------------
+# reconcilePrefix deletes whatever the manifest omits, so an empty manifest is a
+# request to delete the live site. A build that enumerated nothing has failed.
+empty="$tmp/empty"; mkdir -p "$empty"
+plan "$empty" >/dev/null 2>&1
+t "empty export is refused"  "$?"  "1"
+only_cname="$tmp/onlycname"; mkdir -p "$only_cname"; echo x > "$only_cname/CNAME"
+plan "$only_cname" >/dev/null 2>&1
+t "a dir holding only CNAME is empty too"  "$?"  "1"
+plan "$tmp/does-not-exist" >/dev/null 2>&1
+t "missing export dir is refused"  "$?"  "1"
+
+# --- the credential ----------------------------------------------------------
+# Not in PLAN mode (that is the offline seam), but a real run must refuse to
+# start rather than enqueue a deployment it cannot complete.
+out=$(HANZO_DEPLOY_TOKEN= bash "$SD" a-slug "$site" 2>&1); rc=$?
+t "no token: exits non-zero"          "$rc"  "1"
+t "no token: says which secret"       "$(printf '%s' "$out" | grep -c HANZO_DEPLOY_TOKEN)"  "1"
+
+[ $fail -eq 0 ] && echo "PASS" || echo "FAIL"
+exit $fail

bin/sitepublish

@@ -0,0 +1,220 @@
+#!/usr/bin/env bash
+# sitepublish — publish a built static export as an immutable Release on the
+# Hanzo Sites plane. One implementation, every `site:` in the fleet.
+#
+#   sitepublish <slug> <dir>      # e.g. sitepublish hanzo-console out
+#
+# WHY THIS EXISTS: the reusable's `site:` lane used to stage the export with
+# `mc mirror` against hanzoai/s3, which needed S3_ADMIN_ACCESS_KEY and
+# S3_ADMIN_SECRET_KEY. Those names are not in KMS for any org, and the lane
+# fails closed without them — so `site:` refused for every caller that ever
+# declared it, and hanzoai/console had to grow its own build→zip→publish
+# workflow to ship at all. Per-repo build logic is the one thing this repo
+# exists to prevent. The bytes now travel the SAME route console proved, and
+# the credential is the one CI already holds.
+#
+# THE TWO STEPS:
+#
+#   POST /v1/projects/<slug>/deploy   Content-Type: application/zip, zip as body
+#   POST /v1/sites/<slug>/publish     {"source":"<slug>"}
+#
+# The first lands the export at the org's site prefix; the second promotes that
+# prefix into an immutable Release and flips it live. `source` is ORG-RELATIVE
+# and the org segment is prepended server-side from the validated principal
+# (cloud apps/projects/blob.go:60, `sitePrefix(org, slug) = org + "/" + slug`),
+# which is why "<slug>" is the whole of it and why a caller cannot address
+# another tenant's bytes by writing a longer path.
+#
+# ONE CREDENTIAL, ALREADY MINTED. HANZO_API_TOKEN is the IAM JWT the workflow's
+# KMS step already holds: cloud's /v1/kms/auth/login is a broker that performs
+# the IAM client_credentials exchange and returns IAM's own JWT verbatim, so the
+# token that reads this org's secrets is the token that publishes this org's
+# site. api.hanzo.ai validates it and mints X-Org-Id from the `owner` claim —
+# identity is never a header this script sends, so there is nothing here to
+# spoof. Do NOT reintroduce S3_ADMIN_*; a standing bucket key was what this
+# replaced.
+#
+# THE SIZE BOUNDARY IS THE SERVER'S, NOT A PREFERENCE. cloud's public edge sets
+# BodyLimit from GATEWAY_BODY_LIMIT, default 16 MiB (cloud config.go:312), and
+# fasthttp refuses an oversized POST BEFORE any handler runs — it surfaces as an
+# opaque 400 "Error when parsing request" that reads like a malformed payload
+# rather than a size cap (cloud apps/projects/grant.go:5 tells that story about
+# a ~170 MB export). So the zip is measured HERE and refused HERE, with the
+# number and the alternative, instead of being sent to fail unreadably. Measured
+# across the estate's 24 built exports, 22 fit; hanzo.ai (27.9 MiB zipped, 8536
+# files) and trillerfest.com (76.7 MiB) do not. Those go to bin/sitedeploy,
+# which streams per-file against a presigned grant and has no body limit —
+# hanzo.ai is over the server's own 5000-entry cap anyway, so no transport makes
+# it a Release.
+#
+# ENV: HANZO_API_TOKEN  (required) the IAM bearer; the workflow's KMS step mints it
+#      HANZO_API        (https://api.hanzo.ai)
+#      SITEPUBLISH_MAX_ZIP  (16777216) mirror of the server's GATEWAY_BODY_LIMIT
+#      SITEPUBLISH_PLAN=1   print what it would send and exit; no network. Test seam.
+set -euo pipefail
+
+slug="${1:?usage: sitepublish <slug> <dir>}"
+dir="${2:?usage: sitepublish <slug> <dir>}"
+api="${HANZO_API:-https://api.hanzo.ai}"
+
+# Server-side caps, mirrored so a breach is named in CI instead of arriving as a
+# 413 (or, for the body limit, as an unreadable 400). cloud apps/projects/blob.go
+# :29-31. bin/sitepublish_test.sh pins these numbers; if cloud moves them the
+# test is what catches the drift.
+max_zip="${SITEPUBLISH_MAX_ZIP:-16777216}"   # gateway BodyLimit, 16 MiB
+max_files=5000                                # maxFiles
+max_file_bytes=$((64 << 20))                  # maxFileBytes, 64 MiB
+max_total_bytes=$((512 << 20))                # maxTotalBytes uncompressed, 512 MiB
+
+die() { echo "::error::$*"; exit 1; }
+
+# The slug is a URL path segment AND an S3 key segment, so it is held to the
+# grammar cloud validates project slugs with — checked before anything is built
+# or sent, because a bad slug is a typo to fix and not a 404 to interpret.
+echo "$slug" | grep -qE '^[a-z0-9]([a-z0-9-]{0,38}[a-z0-9])?$' \
+  || die "slug '$slug' is not a project slug (^[a-z0-9]([a-z0-9-]{0,38}[a-z0-9])?\$)"
+[ -d "$dir" ] || die "'$dir' is not a directory — the build produced no export"
+
+# A site is a thing with an index. cloud enforces it however the bytes arrive;
+# checking here turns a failed build into a failed publish with a readable cause
+# rather than a promoted release that 404s at its own root.
+[ -f "$dir/index.html" ] || die "'$dir' has no index.html at its root — /v1/sites refuses a source without one"
+
+# CNAME does not travel: a GitHub Pages artifact that means nothing to S3 and
+# would ship a stale hostname claim into the bucket. Same exclusion bin/sitedeploy
+# makes, for the same reason.
+files=$(cd "$dir" && find . -type f ! -name CNAME | wc -l)
+[ "$files" -gt 0 ] || die "'$dir' contains no files — refusing to publish an empty release"
+[ "$files" -le "$max_files" ] \
+  || die "'$dir' holds $files files; cloud caps one artifact at $max_files (apps/projects/blob.go maxFiles). Use bin/sitedeploy — but note the Releases plane has the same cap, so this export cannot become a Release."
+
+bytes=$(cd "$dir" && find . -type f ! -name CNAME -exec wc -c {} + | awk '$2!="total"{s+=$1} END{print s+0}')
+[ "$bytes" -le "$max_total_bytes" ] \
+  || die "'$dir' is $bytes bytes uncompressed; cloud caps an artifact at $max_total_bytes (maxTotalBytes)"
+
+# Per-file cap, checked with find rather than a loop so a 9000-file export costs
+# one traversal. -size uses 512-byte blocks with `c` for bytes; +N c is "strictly
+# greater than N bytes", which is the cap's own boundary.
+big=$(cd "$dir" && find . -type f ! -name CNAME -size +${max_file_bytes}c -printf '%P (%s bytes)\n' | head -3)
+[ -z "$big" ] || die "these files exceed cloud's ${max_file_bytes}-byte per-file cap (maxFileBytes):
+$big"
+
+if [ -n "${SITEPUBLISH_PLAN:-}" ]; then
+  printf 'slug=%s dir=%s files=%s bytes=%s api=%s source=%s\n' \
+    "$slug" "$dir" "$files" "$bytes" "$api" "$slug"
+  exit 0
+fi
+
+# No apostrophe in this message: inside "${VAR:?word}" a single quote opens a
+# quoted section for the PARSER, and the script dies at EOF with "unexpected EOF
+# while looking for matching quote" — a syntax error reported at the last line,
+# nowhere near the one that caused it.
+: "${HANZO_API_TOKEN:?HANZO_API_TOKEN is unset. It is the IAM bearer minted by the KMS step of the reusable workflow, from KMS_CLIENT_ID/KMS_CLIENT_SECRET; there is no second credential to seal.}"
+
+tmp=$(mktemp -d); trap 'rm -rf "$tmp"' EXIT
+zipf="$tmp/site.zip"
+
+# -X drops extra file attributes (uid/gid/timestamps beyond the DOS fields) so
+# the same tree zips to the same bytes on any runner. Entries are stored
+# RELATIVE to the export root — cloud strips a leading "./" and refuses absolute
+# or ".."-escaping names (blob.go:362-370), and strips a common top-level prefix
+# if one exists, so a flat archive is what it expects and what this sends.
+( cd "$dir" && zip -qXr "$zipf" . -x './CNAME' ) || die "zip of '$dir' failed"
+zbytes=$(wc -c < "$zipf")
+
+# The refusal the whole size story is about. Above this the POST dies at the
+# edge with a 400 that names nothing.
+[ "$zbytes" -le "$max_zip" ] || die "the zipped export is $zbytes bytes, past cloud's ${max_zip}-byte edge BodyLimit (GATEWAY_BODY_LIMIT, cloud config.go). fasthttp refuses the POST before any handler runs and reports only 'Error when parsing request', so this is refused here where the number is visible. Publish this export with bin/sitedeploy, which streams per-file against a presigned grant and has no body limit."
+
+echo "publishing $slug — $files files, $bytes bytes ($zbytes zipped)"
+
+# ---- 1. the bytes -----------------------------------------------------------
+# Every response body is KEPT and printed on failure. cloud answers these routes
+# with a specific status per cause — 402 hosting gate, 403 wrong org, 404 no such
+# site for this tenant, 409 the source moved mid-publish, 413 past the caps, 503
+# storage unconfigured — and each is a different fix, so discarding the body
+# discards the answer.
+code=$(curl -sS -o "$tmp/deploy.json" -w '%{http_code}' \
+  -X POST "$api/v1/projects/$slug/deploy" \
+  -H "Authorization: Bearer $HANZO_API_TOKEN" \
+  -H 'Content-Type: application/zip' \
+  --data-binary "@$zipf") \
+  || die "POST $api/v1/projects/$slug/deploy did not complete"
+case "$code" in
+  2??) ;;
+  *) echo "::error::upload $slug → HTTP $code"; head -c 800 "$tmp/deploy.json"; echo; exit 1 ;;
+esac
+
+# ---- 2. promote + activate --------------------------------------------------
+# `source` is the org-relative prefix the upload just landed at. Content types
+# and cache policy are not ours to set: copyRelease rewrites the metadata on
+# every object it promotes, so a release serves identically however its bytes
+# were staged.
+code=$(curl -sS -o "$tmp/publish.json" -w '%{http_code}' \
+  -X POST "$api/v1/sites/$slug/publish" \
+  -H "Authorization: Bearer $HANZO_API_TOKEN" \
+  -H 'Content-Type: application/json' \
+  -d "$(jq -nc --arg s "$slug" '{source:$s}')") \
+  || die "POST $api/v1/sites/$slug/publish did not complete"
+case "$code" in
+  2??) ;;
+  *) echo "::error::publish $slug → HTTP $code"; head -c 800 "$tmp/publish.json"; echo; exit 1 ;;
+esac
+
+# Guard the SHAPE before reading a field out of it. `jq -r '.releaseId // empty'`
+# on an error body, an array, or a null yields "" and every later test on it
+# passes vacuously — which is the defect class this whole script is written
+# against. Assert it is an object carrying the field, then read it.
+jq -e 'type == "object" and (.releaseId | type) == "string" and (.releaseId | length) > 0' \
+  "$tmp/publish.json" >/dev/null \
+  || { echo "::error::publish answered $code but the body is not a release object with a releaseId:"; head -c 800 "$tmp/publish.json"; echo; exit 1; }
+rid=$(jq -r '.releaseId' "$tmp/publish.json")
+objs=$(jq -r '.objects // 0' "$tmp/publish.json")
+url=$(jq -r '.url // ""' "$tmp/publish.json")
+
+# ---- 3. verify the release we just made is the one that is live -------------
+# A 200 from publish says the request was accepted, not that this release is
+# serving. The list is the only thing that can say so.
+#
+# THE LIST IS A BARE JSON ARRAY — `type projectsReleases []projectsRelease`
+# (cloud apps/projects/release.go:518). `.releases[]` against it resolves
+# NOTHING, and an assertion that resolves nothing is an assertion that cannot
+# fail for the reason it was written. Hence `.[]`, and hence the shape guard
+# first: if cloud ever wraps this in an object, THAT is what goes red, loudly,
+# instead of the check quietly matching zero releases forever.
+code=$(curl -sS -o "$tmp/releases.json" -w '%{http_code}' \
+  "$api/v1/sites/$slug/releases" \
+  -H "Authorization: Bearer $HANZO_API_TOKEN") \
+  || die "GET $api/v1/sites/$slug/releases did not complete"
+case "$code" in
+  2??) ;;
+  *) echo "::error::list releases for $slug → HTTP $code"; head -c 800 "$tmp/releases.json"; echo; exit 1 ;;
+esac
+jq -e 'type == "array" and length > 0' "$tmp/releases.json" >/dev/null \
+  || { echo "::error::GET /v1/sites/$slug/releases did not answer a non-empty JSON array (it is the bare-array shape from release.go:518). Body:"; head -c 800 "$tmp/releases.json"; echo; exit 1; }
+
+# Exactly one release is active. Counting first is what makes "none active"
+# distinguishable from "the wrong one is active" — a bare grep for '"active":true'
+# matches ANY release in the list and would pass on both.
+nactive=$(jq '[.[] | select(.active == true)] | length' "$tmp/releases.json")
+[ "$nactive" = 1 ] \
+  || { echo "::error::expected exactly 1 active release for $slug, found $nactive:"; jq -c '[.[]|{releaseId,active}]' "$tmp/releases.json"; exit 1; }
+
+active=$(jq -r 'map(select(.active == true))[0].releaseId // ""' "$tmp/releases.json")
+# Both sides proven non-empty BEFORE they are compared. `[ "$a" = "$b" ]` with
+# two empty strings is TRUE, so an equality test on unguarded values reports
+# success precisely when it learned nothing.
+[ -n "$active" ] || { echo "::error::the active release for $slug carries no releaseId:"; jq -c '.' "$tmp/releases.json"; exit 1; }
+[ -n "$rid" ] || die "publish returned no releaseId to verify against"
+[ "$active" = "$rid" ] \
+  || { echo "::error::published $rid but $active is live for $slug — the flip did not take"; exit 1; }
+
+echo "live: $rid ($objs objects) → ${url:-no public host — that slug is claimed; rename the site or bind a domain}"
+
+if [ -n "${GITHUB_STEP_SUMMARY:-}" ]; then
+  { echo "### Site — \`$slug\`"; echo
+    echo "| release | objects | files | zipped | url |"
+    echo "|---|---|---|---|---|"
+    echo "| \`$rid\` | $objs | $files | $zbytes B | ${url:-—} |"
+  } >> "$GITHUB_STEP_SUMMARY"
+fi

bin/sitepublish_test.sh

@@ -0,0 +1,168 @@
+#!/usr/bin/env bash
+# Tests for bin/sitepublish. Runs OFFLINE and deterministically: a `curl` shim on
+# PATH answers the three routes from fixtures, so every case here — including the
+# ones that must FAIL — needs no token, no bucket and no cluster.
+# Run: bash bin/sitepublish_test.sh
+#
+# The suite is weighted toward REFUSALS on purpose. A publish step that reports
+# success when it verified nothing is worse than one that does not verify at all,
+# because it is indistinguishable from a working one until a site silently stops
+# updating. Three real defects of that exact shape are pinned below by name:
+#
+#   • an object-shaped read (`.releaseId`) against a body that is not an object
+#   • `.releases[]` against the BARE ARRAY /v1/sites/<slug>/releases returns,
+#     which resolves nothing and therefore can never fail for its stated reason
+#   • `[ "$a" = "$b" ]` on two values that are both empty, which is TRUE
+set -uo pipefail
+cd "$(dirname "$0")/.."
+SP="$PWD/bin/sitepublish"
+tmp=$(mktemp -d); trap 'rm -rf "$tmp"' EXIT
+fail=0
+
+t() { # t <name> <want_rc> <got_rc> [<must-contain> <output>]
+  local name="$1" want="$2" got="$3" needle="${4:-}" out="${5:-}"
+  if [ "$got" != "$want" ]; then
+    printf 'FAIL  %-58s rc=%s (want %s)\n' "$name" "$got" "$want"; fail=1; return
+  fi
+  if [ -n "$needle" ] && ! printf '%s' "$out" | grep -qF -- "$needle"; then
+    printf 'FAIL  %-58s rc=%s but missing %q\n' "$name" "$got" "$needle"; fail=1
+    printf '      got: %s\n' "$(printf '%s' "$out" | head -c 300)"; return
+  fi
+  printf 'ok    %-58s rc=%s\n' "$name" "$got"
+}
+
+# ---- a curl shim -------------------------------------------------------------
+# Dispatches on the URL and writes the fixture the scenario names to the file the
+# real curl would have written, then prints the status the way `-w %{http_code}`
+# does. Everything the script does with a response goes through this, so the
+# tests exercise the REAL parsing, not a mock of it.
+shim="$tmp/bin"; mkdir -p "$shim"
+cat > "$shim/curl" <<'SHIM'
+#!/usr/bin/env bash
+out=/dev/null url=
+while [ $# -gt 0 ]; do
+  case "$1" in
+    -o) out="$2"; shift 2 ;;
+    http*|https*) url="$1"; shift ;;
+    *) shift ;;
+  esac
+done
+case "$url" in
+  */deploy)   printf '%s' "${T_DEPLOY_BODY:-{\"ok\":true\}}" > "$out"; printf '%s' "${T_DEPLOY_CODE:-200}" ;;
+  */publish)  printf '%s' "${T_PUBLISH_BODY:-}"              > "$out"; printf '%s' "${T_PUBLISH_CODE:-200}" ;;
+  */releases) printf '%s' "${T_LIST_BODY:-}"                 > "$out"; printf '%s' "${T_LIST_CODE:-200}" ;;
+  *) echo "shim: unexpected url $url" >&2; exit 9 ;;
+esac
+SHIM
+chmod +x "$shim/curl"
+
+site="$tmp/site"; mkdir -p "$site/assets"
+echo '<h1>hi</h1>' > "$site/index.html"
+echo 'body{}'      > "$site/assets/app.css"
+echo 'hanzo.ai'    > "$site/CNAME"
+
+run() { # run <dir> — publish with the shim on PATH, current T_* scenario
+  PATH="$shim:$PATH" HANZO_API_TOKEN=tok HANZO_API=https://api.test \
+    bash "$SP" a-slug "$1" 2>&1
+}
+OK_PUB='{"releaseId":"rel-1","slug":"a-slug","objects":2,"active":true,"url":"https://a-slug.hanzo.page"}'
+OK_LIST='[{"releaseId":"rel-1","active":true},{"releaseId":"rel-0","active":false}]'
+
+# ---- the plan seam: no network, no token ------------------------------------
+out=$(SITEPUBLISH_PLAN=1 bash "$SP" a-slug "$site" 2>&1); rc=$?
+t "plan prints the manifest and exits" 0 $rc "files=2" "$out"
+# CNAME is a GitHub Pages artifact: it means nothing to S3 and would ship a stale
+# hostname claim. Two files, not three.
+t "plan excludes CNAME from the count" 0 $rc "files=2 " "$out"
+t "plan sends source=<slug>, org-relative" 0 $rc "source=a-slug" "$out"
+
+# ---- refusals that need no network ------------------------------------------
+out=$(SITEPUBLISH_PLAN=1 bash "$SP" 'Bad_Slug' "$site" 2>&1); rc=$?
+t "an invalid slug is refused before anything" 1 $rc "is not a project slug" "$out"
+out=$(SITEPUBLISH_PLAN=1 bash "$SP" a-slug "$tmp/nope" 2>&1); rc=$?
+t "a missing export dir is refused" 1 $rc "is not a directory" "$out"
+noidx="$tmp/noidx"; mkdir -p "$noidx"; echo x > "$noidx/page.html"
+out=$(SITEPUBLISH_PLAN=1 bash "$SP" a-slug "$noidx" 2>&1); rc=$?
+t "an export with no index.html is refused" 1 $rc "no index.html at its root" "$out"
+empty="$tmp/empty"; mkdir -p "$empty"; echo x > "$empty/index.html"; rm "$empty/index.html"
+out=$(SITEPUBLISH_PLAN=1 bash "$SP" a-slug "$empty" 2>&1); rc=$?
+t "an empty export is refused" 1 $rc "no index.html" "$out"
+
+# The size refusal is the whole reason this script measures before it sends:
+# past the edge BodyLimit fasthttp rejects the POST before any handler runs and
+# reports only "Error when parsing request", which names neither size nor cause.
+out=$(SITEPUBLISH_MAX_ZIP=1 PATH="$shim:$PATH" HANZO_API_TOKEN=tok \
+      bash "$SP" a-slug "$site" 2>&1); rc=$?
+t "a zip past the edge BodyLimit is refused here" 1 $rc "BodyLimit" "$out"
+t "  ...and the refusal names bin/sitedeploy" 1 $rc "bin/sitedeploy" "$out"
+
+# The server caps one artifact at 5000 entries; a 5001-file export can never
+# become a Release by ANY transport, so saying so here beats a 413 later.
+many="$tmp/many"; mkdir -p "$many"; echo x > "$many/index.html"
+( cd "$many" && touch f{1..5001} )
+out=$(SITEPUBLISH_PLAN=1 bash "$SP" a-slug "$many" 2>&1); rc=$?
+t "an export past maxFiles=5000 is refused" 1 $rc "cloud caps one artifact at 5000" "$out"
+
+# ---- the happy path ----------------------------------------------------------
+out=$(T_PUBLISH_BODY="$OK_PUB" T_LIST_BODY="$OK_LIST" run "$site"); rc=$?
+t "publish + verified flip succeeds" 0 $rc "live: rel-1" "$out"
+
+# ---- transport failures carry the server's answer ---------------------------
+# Each status is a different fix, so the body is printed rather than swallowed.
+out=$(T_DEPLOY_CODE=413 T_DEPLOY_BODY='{"error":"artifact exceeds"}' run "$site"); rc=$?
+t "a non-2xx upload fails and prints the body" 1 $rc "artifact exceeds" "$out"
+out=$(T_PUBLISH_CODE=402 T_PUBLISH_BODY='{"error":"hosting not enabled"}' run "$site"); rc=$?
+t "a non-2xx publish fails and prints the body" 1 $rc "hosting not enabled" "$out"
+out=$(T_PUBLISH_BODY="$OK_PUB" T_LIST_CODE=503 T_LIST_BODY='{"error":"storage"}' run "$site"); rc=$?
+t "a non-2xx release list fails" 1 $rc "list releases" "$out"
+
+# ---- defect 1: an object-shaped read against a non-object -------------------
+# `jq -r '.releaseId // empty'` yields "" for an array, a null, or an error body,
+# and every later test on "" passes vacuously. The shape is asserted first.
+out=$(T_PUBLISH_BODY='[{"releaseId":"rel-1"}]' T_LIST_BODY="$OK_LIST" run "$site"); rc=$?
+t "publish answering an ARRAY is refused" 1 $rc "not a release object" "$out"
+out=$(T_PUBLISH_BODY='{"ok":true}' T_LIST_BODY="$OK_LIST" run "$site"); rc=$?
+t "publish answering no releaseId is refused" 1 $rc "not a release object" "$out"
+out=$(T_PUBLISH_BODY='{"releaseId":""}' T_LIST_BODY="$OK_LIST" run "$site"); rc=$?
+t "publish answering an EMPTY releaseId is refused" 1 $rc "not a release object" "$out"
+out=$(T_PUBLISH_BODY='not json' T_LIST_BODY="$OK_LIST" run "$site"); rc=$?
+t "publish answering non-JSON is refused" 1 $rc "not a release object" "$out"
+
+# ---- defect 2: the list is a BARE ARRAY -------------------------------------
+# release.go:518 is `type projectsReleases []projectsRelease`. A `.releases[]`
+# filter resolves NOTHING against that, so an assertion built on it can only ever
+# refuse — it never once tested what it claimed to. The guard demands the array
+# shape, so the day cloud wraps the list THAT is what goes red, by name.
+out=$(T_PUBLISH_BODY="$OK_PUB" T_LIST_BODY='{"releases":[{"releaseId":"rel-1","active":true}]}' run "$site"); rc=$?
+t "a WRAPPED release list is refused, loudly" 1 $rc "bare-array shape" "$out"
+out=$(T_PUBLISH_BODY="$OK_PUB" T_LIST_BODY='[]' run "$site"); rc=$?
+t "an EMPTY release list is refused" 1 $rc "non-empty JSON array" "$out"
+out=$(T_PUBLISH_BODY="$OK_PUB" T_LIST_BODY='null' run "$site"); rc=$?
+t "a null release list is refused" 1 $rc "non-empty JSON array" "$out"
+
+# ---- defect 3: counting, so "none" cannot read as "yes" ---------------------
+# A bare `grep '"active":true'` matches ANY release in the list, so it passes
+# both when the wrong release is live and when the list merely mentions one.
+out=$(T_PUBLISH_BODY="$OK_PUB" T_LIST_BODY='[{"releaseId":"rel-1","active":false}]' run "$site"); rc=$?
+t "ZERO active releases is refused" 1 $rc "expected exactly 1 active release" "$out"
+out=$(T_PUBLISH_BODY="$OK_PUB" \
+      T_LIST_BODY='[{"releaseId":"rel-1","active":true},{"releaseId":"rel-2","active":true}]' run "$site"); rc=$?
+t "TWO active releases is refused" 1 $rc "expected exactly 1 active release" "$out"
+out=$(T_PUBLISH_BODY="$OK_PUB" T_LIST_BODY='[{"releaseId":"rel-9","active":true}]' run "$site"); rc=$?
+t "a DIFFERENT release being live is refused" 1 $rc "the flip did not take" "$out"
+
+# The both-empty comparison, head on: an active entry whose releaseId is "".
+# `[ "$active" = "$rid" ]` with both empty is TRUE and would report success on a
+# release that does not exist. Non-emptiness is proven before the comparison.
+out=$(T_PUBLISH_BODY="$OK_PUB" T_LIST_BODY='[{"releaseId":"","active":true}]' run "$site"); rc=$?
+t "an active release with an EMPTY id is refused" 1 $rc "carries no releaseId" "$out"
+out=$(T_PUBLISH_BODY="$OK_PUB" T_LIST_BODY='[{"active":true}]' run "$site"); rc=$?
+t "an active release with NO id field is refused" 1 $rc "carries no releaseId" "$out"
+
+# ---- the credential is required ---------------------------------------------
+out=$(PATH="$shim:$PATH" HANZO_API_TOKEN= bash "$SP" a-slug "$site" 2>&1); rc=$?
+t "a missing bearer refuses before any request" 1 $rc "HANZO_API_TOKEN is unset" "$out"
+
+echo
+[ "$fail" = 0 ] && echo "sitepublish: all tests passed" || echo "sitepublish: FAILURES"
+exit "$fail"

bin/vendormark

@@ -0,0 +1,161 @@
+#!/usr/bin/env bash
+# vendormark — refuse a repo that carries a site-generator vendor's branding.
+# One implementation, every caller.
+#
+#   vendormark [dir] [git-range]
+#
+# WHAT THIS CATCHES
+#
+# Several sites in the estate began life as output from a hosted "prompt to
+# React app" generator. The generator does not sign its work in a comment you
+# would notice — it signs it in the SHAPE of the repo, in four places at once:
+#
+#   1. an asset folder named after the vendor, e.g. public/<vendor>-uploads/,
+#      so every image the browser requests carries the vendor's name in its URL
+#   2. a build plugin dependency in package.json and the lockfiles
+#   3. an injected third-party <script> tag in index.html
+#   4. commit messages, author identities and Co-authored-by trailers written
+#      by the vendor's bot
+#
+# The fourth is the one nothing else looks at, and the one that survives every
+# amount of tidying the working tree: a repo can render a perfect page and
+# still say, in `git log`, who really wrote it.
+#
+# This was not a hypothetical. A sweep of 1,065 repositories across the three
+# orgs found the mark in five shipped sites — one of them serving the vendor's
+# upload folder as its FAVICON, so the vendor's path was in the <head> of every
+# page and in every browser tab. Purging it needed a history rewrite and a
+# force-push of every branch. This gate exists so that is never needed twice.
+#
+# WHY THE CONTENT RULES ARE SO SPECIFIC
+#
+# The vendor's name is also an ordinary English adjective, and the naive gate —
+# case-insensitive substring of that word — is unshippable. Measured across all
+# 1,065 repos, the bare word appears legitimately in 20+ of them and NONE of
+# them are contaminated:
+#
+#     tokenizer vocab.json (3 repos)     "lovable</w>": 38565
+#     English word lists                 words_alpha.txt, faker adjectives
+#     ML training/eval corpora           alpaca, kto, BAGEL eval prompts
+#     x/net publicsuffix packed data     a substring of concatenated domains
+#     vendored prompt corpora            third-party collections that quote it
+#     our own marketing copy             a competitor named in a case study
+#
+# A gate that reds a fifth of the estate on day one is a gate someone switches
+# off, and then we are worse off than before it existed. So the content rules
+# are only the strings that CANNOT occur in prose: a path segment, an npm
+# package name, an injected script URL, a generator meta tag, a bot's email.
+#
+# Two candidate rules were DROPPED after measuring them, and it is worth saying
+# why: `lovableproject.com` and `lovable.app` red an upstream Clerk SDK bundle,
+# which ships a list of dev-preview host suffixes naming several generators. A
+# third party's honest list of other people's hosts is not our contamination.
+# Those two strings are still refused in commit messages and git identities,
+# where no third party can put them.
+#
+# PATHS AND IDENTITIES ARE ABSOLUTE
+#
+# Content is judged narrowly; paths and git identities are not. There is no
+# legitimate reason for the vendor's name to appear in a filename we track or
+# in an author, committer or Co-authored-by line we write. Those are refused on
+# the bare word, no exceptions, because that is where the mark is load-bearing
+# and where tidying the working tree never reaches.
+#
+# THE ESCAPE HATCH
+#
+# `.vendormark-allow` — one path glob per line, `#` comments ignored. It exists
+# for exactly one honest case: a repo whose JOB is to name these vendors, such
+# as the history-scrubber rule files in hanzoai/.github, which cannot do their
+# work without spelling the string they remove. If you are reaching for it for
+# any other reason, you are about to ship the mark.
+#
+# EXIT: 0 clean, 1 mark found.
+set -uo pipefail
+
+root=${1:-.}
+range=${2:-}
+cd "$root" 2>/dev/null || { echo "vendormark: no such directory: $root" >&2; exit 1; }
+git rev-parse --git-dir >/dev/null 2>&1 || exit 0   # not a repo: nothing to check
+
+# PATHS: the bare vendor name, no exceptions. Add a vendor here and its
+# filenames are refused everywhere.
+VENDORS='lovable'
+
+# CONTENT: only strings that cannot occur in prose — an asset folder, an npm
+# package, the injected script, a generator meta tag, the bot's address. Spelled
+# out in full rather than composed from $VENDORS, because half of them do not
+# contain the vendor's name at all. See the header for the two that were
+# measured against the estate and dropped.
+CONTENT_SIG='lovable-uploads/|lovable-tagger|@lovable\.dev|cdn\.gpteng\.co|gptengineer\.js|content="[Ll]ovable"'
+
+# IDENTITIES AND MESSAGES: the bare word plus the vendor hosts, because nothing
+# outside our own commits can write these.
+IDENT_SIG='lovable|gptengineer|gpt-engineer'
+
+fail=0
+tmp=$(mktemp -d); trap 'rm -rf "$tmp"' EXIT
+
+# --- allowlist ------------------------------------------------------------
+: > "$tmp/allow"
+if [ -f .vendormark-allow ]; then
+  grep -vE '^\s*(#|$)' .vendormark-allow > "$tmp/allow" || true
+fi
+allowed() {  # $1 = path
+  [ -s "$tmp/allow" ] || return 1
+  while IFS= read -r g; do
+    # shellcheck disable=SC2254
+    case "$1" in $g) return 0;; esac
+  done < "$tmp/allow"
+  return 1
+}
+
+# --- 1. tracked paths -----------------------------------------------------
+git ls-files 2>/dev/null | grep -iE "$VENDORS" > "$tmp/paths" || true
+if [ -s "$tmp/paths" ]; then
+  n=0
+  while IFS= read -r p; do allowed "$p" || { echo "  path:    $p"; n=$((n+1)); }; done < "$tmp/paths"
+  if [ "$n" -gt 0 ]; then
+    echo "::error::${n} tracked path(s) carry a site-generator vendor's name. Every one of them is a URL the browser requests, so the vendor's branding is in the page. Move the assets to a neutral folder (public/img/) and repoint the references in the same change — a favicon or logo that a <link> or <img> depends on must be REPLACED with the real brand mark, not merely deleted."
+    fail=1
+  fi
+fi
+
+# --- 2. tracked content ---------------------------------------------------
+git grep -Iin -E "$CONTENT_SIG" HEAD 2>/dev/null | sed 's#^HEAD:##' > "$tmp/content" || true
+if [ -s "$tmp/content" ]; then
+  n=0
+  while IFS= read -r line; do
+    p=${line%%:*}
+    allowed "$p" || { echo "  content: $(echo "$line" | cut -c1-140)"; n=$((n+1)); }
+  done < "$tmp/content"
+  if [ "$n" -gt 0 ]; then
+    echo "::error::${n} tracked line(s) reference a site-generator vendor's asset folder, build plugin or injected script. Drop the dependency and the script tag; repoint the asset URLs."
+    fail=1
+  fi
+fi
+
+# --- 3. commit messages, identities, trailers -----------------------------
+# Default scope is HEAD alone, which is what a fetch-depth:1 CI checkout has and
+# is enough to refuse the commit being pushed. Pass a range (or set
+# VENDORMARK_RANGE) to judge more, e.g. origin/main..HEAD for a whole PR.
+range=${range:-${VENDORMARK_RANGE:-}}
+if [ -n "$range" ]; then set -- "$range"; else set -- -1 HEAD; fi
+git log "$@" --format='%H%x01%an <%ae>%x01%cn <%ce>%x01%B%x01%(trailers:unfold=true)%x02' 2>/dev/null \
+  | tr -d '\n' | tr '\002' '\n' > "$tmp/log" || true
+if [ -s "$tmp/log" ]; then
+  n=0
+  while IFS= read -r c; do
+    echo "$c" | grep -qiE "$IDENT_SIG" || continue
+    echo "  commit:  $(echo "$c" | tr '\001' ' ' | cut -c1-150)"
+    n=$((n+1))
+  done < "$tmp/log"
+  if [ "$n" -gt 0 ]; then
+    echo "::error::${n} commit(s) name a site-generator vendor in the message, the author/committer identity or a Co-authored-by trailer. A working tree can be tidied; \`git log\` cannot be, short of a history rewrite and a force-push of every branch. Reword before pushing. Describe the work — \"generated placeholder assets\" — and never credit the generator."
+    fail=1
+  fi
+fi
+
+if [ "$fail" -eq 0 ]; then
+  echo "vendormark: OK — no site-generator branding in paths, tracked content, messages or identities"
+fi
+exit "$fail"

bin/vendormark_test.sh

@@ -0,0 +1,142 @@
+#!/usr/bin/env bash
+# vendormark_test.sh — the refusals AND the allowances, both pinned.
+#
+# A gate is only worth having if it is exact in both directions. A false
+# negative lets the vendor's mark back into a shipped site; a false positive
+# reds a repo that was always fine, and a gate that reds honest repos is a gate
+# someone switches off. So this suite asserts BOTH halves, and the allowance
+# half is the larger one on purpose: the vendor's name is also an ordinary
+# English adjective, and it occurs innocently in tokenizer vocabularies, word
+# lists, ML corpora, packed public-suffix data and third-party host lists all
+# over the estate.
+#
+# Offline and deterministic: temp repos, no network.
+set -uo pipefail
+BIN=$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)/vendormark
+pass=0; fail=0
+V=$(printf 'lo%s' 'vable')   # not spelled literally, so this file does not
+                             # itself trip the gate it is testing
+
+newrepo() {
+  d=$(mktemp -d)
+  git -C "$d" init -q
+  git -C "$d" config user.email dev@hanzo.ai
+  git -C "$d" config user.name "Hanzo Dev"
+  printf 'x\n' > "$d/README.md"
+  git -C "$d" add -A
+  git -C "$d" commit -qm "initial"
+  echo "$d"
+}
+check() { # name expected_rc dir [range]
+  local name=$1 want=$2 dir=$3 range=${4:-}
+  "$BIN" "$dir" $range >/dev/null 2>&1; local got=$?
+  if [ "$got" = "$want" ]; then pass=$((pass+1)); echo "  ok    $name"
+  else fail=$((fail+1)); echo "  FAIL  $name (want rc=$want, got rc=$got)"; fi
+}
+
+echo "REFUSALS — the mark in each of the four places it hides"
+
+d=$(newrepo)
+check "clean repo is green" 0 "$d"
+mkdir -p "$d/public/$V-uploads"
+printf 'PNG\n' > "$d/public/$V-uploads/28d53ec4.png"
+git -C "$d" add -A && git -C "$d" commit -qm "add an image"
+check "MUTATION: tracked path named for the vendor" 1 "$d"
+git -C "$d" rm -rq "public/$V-uploads" && git -C "$d" commit -qm "move the image"
+check "MUTATION REVERTED: green again" 0 "$d"
+rm -rf "$d"
+
+d=$(newrepo)
+printf '<img src="/%s-uploads/28d53ec4.png" />\n' "$V" > "$d/index.html"
+git -C "$d" add -A && git -C "$d" commit -qm "a page"
+check "content: asset folder referenced in markup" 1 "$d"
+rm -rf "$d"
+
+d=$(newrepo)
+printf '{"devDependencies":{"%s-tagger":"^1.1.3"}}\n' "$V" > "$d/package.json"
+git -C "$d" add -A && git -C "$d" commit -qm "deps"
+check "content: the vendor's build plugin in package.json" 1 "$d"
+rm -rf "$d"
+
+d=$(newrepo)
+printf '<script src="https://cdn.gpteng.co/gptengineer.js"></script>\n' > "$d/index.html"
+git -C "$d" add -A && git -C "$d" commit -qm "a page"
+check "content: the vendor's injected script tag" 1 "$d"
+rm -rf "$d"
+
+d=$(newrepo)
+printf '<meta name="generator" content="Lovable" />\n' > "$d/index.html"
+git -C "$d" add -A && git -C "$d" commit -qm "a page"
+check "content: the generator meta tag" 1 "$d"
+rm -rf "$d"
+
+d=$(newrepo)
+printf 'y\n' > "$d/a.txt"; git -C "$d" add -A
+git -C "$d" commit -qm "Update $V project template"
+check "MUTATION: the vendor named in a commit message" 1 "$d"
+git -C "$d" commit -q --amend -m "update the project template"
+check "MUTATION REVERTED: reworded message is green" 0 "$d"
+rm -rf "$d"
+
+d=$(newrepo)
+printf 'y\n' > "$d/a.txt"; git -C "$d" add -A
+# --no-verify ON PURPOSE. A commit-msg hook on the workstation
+# (~/.githooks/commit-msg) already strips vendor co-author trailers, and with it
+# enabled this case cannot be constructed — the hook silently rewrites the
+# trailer to ours and the assertion passes for the wrong reason. That hook is
+# per-machine: it does not run on a runner, on a teammate's laptop, or on a
+# commit made through the GitHub web UI. This gate is the layer that does. So
+# the test bypasses the hook to prove the GATE catches what the hook would have.
+git -C "$d" commit -q --no-verify -m "a change
+
+Co-authored-by: $V bot <bot@$V.dev>"
+check "MUTATION: Co-authored-by credits the vendor" 1 "$d"
+git -C "$d" commit -q --amend --no-verify -m "a change"
+check "MUTATION REVERTED: trailer dropped is green" 0 "$d"
+rm -rf "$d"
+
+d=$(newrepo)
+printf 'y\n' > "$d/a.txt"; git -C "$d" add -A
+git -C "$d" -c user.name="$V" -c user.email="bot@$V.dev" commit -qm "a change"
+check "MUTATION: the vendor as author identity" 1 "$d"
+rm -rf "$d"
+
+d=$(newrepo)
+printf 'y\n' > "$d/a.txt"; git -C "$d" add -A; git -C "$d" commit -qm "clean subject"
+printf 'z\n' > "$d/b.txt"; git -C "$d" add -A; git -C "$d" commit -qm "Visual edit in $V"
+printf 'w\n' > "$d/c.txt"; git -C "$d" add -A; git -C "$d" commit -qm "clean again"
+check "default scope (HEAD only) misses an older bad message" 0 "$d"
+check "explicit range catches it" 1 "$d" "HEAD~3..HEAD"
+rm -rf "$d"
+
+echo
+echo "ALLOWANCES — the same word, innocently, as it really occurs in the estate"
+
+d=$(newrepo)
+printf '  "%s</w>": 38565,\n' "$V" > "$d/vocab.json"
+printf '%s\n' "$V" > "$d/words_alpha.txt"
+printf 'Tell me a story about a %s character.\n' "$V" > "$d/harmless.txt"
+printf 'caseStudy: %s and v0 activate users by turning a prompt into an app\n' "${V^}" > "$d/guide.yaml"
+printf 'mail2%s.com\n' "$V" > "$d/generic_emails.txt"
+printf 'let e=[".%s.app",".%sproject.com",".webcontainer-api.io"];\n' "$V" "$V" > "$d/clerk-bundle.js"
+printf 'A collection of UI components. Integrate them in v0, %s, Bolt.\n' "${V^}" > "$d/registries.json"
+git -C "$d" add -A && git -C "$d" commit -qm "corpora, word lists and third-party bundles"
+check "tokenizer vocab / word list / ML corpus / marketing prose / blocklist / upstream host list" 0 "$d"
+rm -rf "$d"
+
+echo
+echo "ESCAPE HATCH — a repo whose job is to name the vendor"
+
+d=$(newrepo)
+mkdir -p "$d/tools/rules"
+printf 'Hanzo Dev <dev@hanzo.ai> <bot@%s.dev>\n' "$V" > "$d/tools/rules/mailmap.txt"
+git -C "$d" add -A && git -C "$d" commit -qm "scrubber rules"
+check "scrubber rule file is refused by default" 1 "$d"
+printf 'tools/rules/*\n' > "$d/.vendormark-allow"
+git -C "$d" add -A && git -C "$d" commit -qm "declare the rule files"
+check "...and allowed once declared in .vendormark-allow" 0 "$d"
+rm -rf "$d"
+
+echo
+echo "vendormark_test: $pass passed, $fail failed"
+[ "$fail" -eq 0 ]

hanzo.yml

@@ -59,6 +59,60 @@ test:
         exit 1
       fi
       echo "OK: .github and .hanzo copies are one file ($(wc -l < .github/workflows/build.yml) lines)"
+  - name: gate-runs-before-delegate
+    # `mode: delegate` chooses WHERE the image is built. It must never choose
+    # WHETHER the commit was tested. It did: the delegate step was evaluated
+    # fourth, ahead of every toolchain and ahead of `test:`, and every other step
+    # carried `if: inputs.mode != 'delegate'` — so a delegated run skipped the
+    # gate along with the build and finished green in seconds having proven
+    # nothing. The seam was right and the ORDER was wrong, which is the kind of
+    # defect that reads as correct in every diff that touches one step at a time.
+    #
+    # Two assertions, because either one alone can be satisfied while the gate
+    # still does not run: the delegate POST must come AFTER the test gate, and
+    # the test gate must not be delegate-guarded.
+    run: |
+      set -e
+      f=.github/workflows/build.yml
+      t=$(grep -n '^      - name: Test (per hanzo.yml)$'                       "$f" | cut -d: -f1)
+      d=$(grep -n '^      - name: Delegate build to platform (mode=delegate)$' "$f" | cut -d: -f1)
+      [ -n "$t" ] && [ -n "$d" ] || { echo "::error::cannot find the test gate ($t) or the delegate step ($d) — a rename broke this assertion, fix the assertion"; exit 1; }
+      [ "$d" -gt "$t" ] || { echo "::error::the delegate step (line $d) runs BEFORE the test gate (line $t) — a delegated build would ship an ungated commit"; exit 1; }
+      if sed -n "${t}p;$((t+1)),$((t+80))p" "$f" | grep -m1 '^        if: ' | grep -q "mode != 'delegate'"; then
+        echo "::error::the test gate is guarded by \`inputs.mode != 'delegate'\` — delegate would skip it"; exit 1
+      fi
+      echo "OK: test gate at line $t, delegate POST at line $d, gate not delegate-guarded"
+  - name: publishable
+    # bin/publishable refuses a `build_secrets` name that has not declared itself
+    # public. That matters because a build_secret is baked in as a --build-arg,
+    # which `docker history` prints to anyone who can pull — so the key is a
+    # publishing decision wearing the word "secret". Until this, the only check
+    # in the fleet lived inside hanzoai/ui's own Dockerfile, for one image.
+    run: bash bin/publishable_test.sh
+  - name: publishable-rule-is-one-rule
+    # The rule is spelled twice on purpose, and this is what keeps it one rule.
+    #
+    # build.yml cannot call bin/publishable, because the tools checkout and the
+    # workflow resolve from DIFFERENT places: a caller pins the workflow at a ref
+    # its own forge resolves, while the tools step derives its ref from
+    # GITHUB_WORKFLOW_REF — which git.hanzo.ai does not set, so it falls back to
+    # `v1` and clones github.com. Measured on run 36473: `derived ref=v1`,
+    # `cloned https://github.com/hanzoai/ci@v1`. So a step calling a NEW file
+    # under $CI_HOME is broken on every forge run until a tag moves on a host the
+    # caller never named. The pipeline has to carry its own rule.
+    #
+    # Which leaves the pattern in two files, which is the same shape as the two
+    # copies of build.yml and gets the same treatment: demand they are identical
+    # and let the gate, not a reader, be the thing that notices.
+    run: |
+      set -e
+      a=$(grep -oE '^[[:space:]]*PUBLISHABLE_\*\|[^)]*\)' .github/workflows/build.yml | tr -d '[:space:]')
+      b=$(grep -oE '^[[:space:]]*PUBLISHABLE_\*\|[^)]*\)' bin/publishable          | tr -d '[:space:]')
+      [ -n "$a" ] && [ -n "$b" ] || { echo "::error::publishable rule not found in build.yml ($a) or bin/publishable ($b) — an edit renamed it, fix this assertion"; exit 1; }
+      [ "$a" = "$b" ] || { echo "::error::the publishable rule differs between build.yml and bin/publishable — build.yml is what runs, bin/publishable is what is tested, so a difference means the tested rule is not the enforced one:
+        build.yml:       $a
+        bin/publishable: $b"; exit 1; }
+      echo "OK: one publishable rule, two spellings — $a"
   - name: imgver
     # bin/imgver decides the version EVERY image in the fleet publishes — this
     # workflow's build lane calls it, and so does the imgver composite action the
@@ -67,6 +121,24 @@ test:
     # IfNotPresent never picks up and no reader can see. Offline and
     # deterministic: the registry floor is injected, so it needs no network.
     run: bash bin/imgver_test.sh
+  - name: vendormark
+    # bin/vendormark is the gate the build lane runs against every checkout: it
+    # refuses a site-generator vendor's branding in a tracked path, in tracked
+    # content, in a commit message, in an author identity or in a Co-authored-by
+    # trailer. Five shipped sites carried it — one serving the vendor's upload
+    # folder as its favicon — and clearing them took a history rewrite and a
+    # force-push of every branch.
+    #
+    # The suite pins the refusals AND the allowances, and the allowances are the
+    # larger half on purpose. The vendor's name is also an ordinary English
+    # adjective: across the 1,065 repos swept it appears innocently in tokenizer
+    # vocabularies, word lists, ML corpora, packed public-suffix data and an
+    # upstream Clerk bundle's list of other people's dev hosts. A gate that reds
+    # a fifth of the estate is a gate someone switches off, so the content rules
+    # are only strings that cannot occur in prose, and two candidate rules were
+    # measured and DROPPED for redding that Clerk bundle. Offline and
+    # deterministic: temp repos, no network.
+    run: bash bin/vendormark_test.sh
   - name: gover
     # bin/gover is the gate the build lane runs against every Dockerfile before
     # it builds: it refuses a Go builder image older than the go.mod it
@@ -83,6 +155,22 @@ test:
     # and a multi-module repo is judged by its NEAREST go.mod. Offline and
     # deterministic: temp dirs, no registry, no network.
     run: bash bin/gover_test.sh
+  - name: sitepublish
+    # bin/sitepublish is the whole of the `site:` lane. The lane it replaced was
+    # dead: it staged with `mc mirror` behind S3_ADMIN_ACCESS_KEY/SECRET_KEY,
+    # names KMS holds for no org, so it failed closed for every caller — and of
+    # the 155 hanzo.yml files across the three orgs, not one declared a `site:`.
+    #
+    # The suite is weighted toward REFUSALS because the failure that matters here
+    # is a green publish that verified nothing — indistinguishable from a working
+    # one until a site quietly stops updating. Three defects of exactly that
+    # shape are pinned by name: an object-shaped read of a body that is not an
+    # object, `.releases[]` against the BARE ARRAY the list route returns (which
+    # resolves nothing and so can never fail for its stated reason), and
+    # `[ "$a" = "$b" ]` on two empty strings, which is TRUE. Reintroducing them
+    # turns 3, 9 and 1 of these tests red respectively. Offline and
+    # deterministic: a curl shim answers the routes from fixtures, no network.
+    run: bash bin/sitepublish_test.sh
 
 # No `deploy:` ON PURPOSE. Rollout is a reviewed tag pin in hanzoai/universe
 # (infra/k8s/operator/crs/ci.yaml), the same rule cloud and git follow: a