Skip to content

feat(site,docs): the demo is answering again, so point at it again - #45

Merged
blakebauman merged 1 commit into
mainfrom
restore-demo-pointers
Aug 26, 2026
Merged

blakebauman merged 1 commit into
mainfrom
restore-demo-pointers

Conversation

@blakebauman

Copy link
Copy Markdown
Member

Reverses #44 for the demo half only. That PR removed the pointers because the thing they pointed at had been deleted in the 2026-08-18 teardown; the gateways are redeployed, so the pointers come back.

Redeployed first

All three workers had been deleted from the account, not idled, so these were fresh creates — new Durable Object namespaces and new container applications.

Worker Hostname Health Container
fold-demo-tasks tasks.fold.run 200 lite, max 1
fold-demo demo.fold.run 200 basic, max 1
fold-enterprise enterprise.fold.run 200 basic, max 1

Restored

  • The homepage #demo section, its three upstream cards, and its styles
  • Nav Live demo; footer Live demo and Live console
  • docs.fold.run/try-the-demo/ and its sidebar entry
  • The benchmarks bullet warning against benchmarking the demo

Deliberately not restored

The /status/ page, the footer status badge and its uptime fetch, and fold-uptime stays undeployed.

The monitor watched the fold cloud as well as the demo, and the fold cloud is still down — api, broker, mcp and console all answer 522. A status page fed by that monitor would come up mostly red and argue against the demo sitting next to it, which is the same failure mode #44 was fixing, pointed the other way. It comes back when there is a cloud for it to report on.

The two sentences that sent readers to /status for live state are dropped rather than left dangling: one in the homepage demo intro, one in the walkthrough's opening line.

Verification

The walkthrough was checked against the running gateway rather than assumed from the restored text:

  • tools/list returns 8 tools across cfdocs (2), git (5), jobs (1), with no run.fold/partialFailure — all three upstreams federate
  • jobs__start_job mints a task into the result's _meta, origin-tagged run.fold/upstream: demo-tasks
  • That task id polls successfully via tasks/get from a brand-new session that never saw it minted, which is the cross-session affinity claim the page makes

Build, typecheck and lint pass. The single lint warning is pre-existing on a clean tree.

Follow-up, not in this PR

Deleting fold-enterprise took its secrets with it, so enterprise.fold.run/demo-token returns 503 demo token vending is not configured yet. The four oauth.work client-credentials values need restoring with wrangler secret put. Nothing on the site links to that endpoint, so it does not block this.

🤖 Generated with Claude Code

The demo gateways were redeployed today — fold-demo-tasks, fold-demo and
fold-enterprise are back on tasks/demo/enterprise.fold.run, all three
healthy, with the walkthrough verified end to end against the running
gateway (three namespaces federate, a minted task polls from a session
that never saw it minted). This restores what #44 removed once the thing
it pointed at was gone: the homepage demo section and its styles, the
nav and footer demo links, the docs walkthrough and its sidebar entry,
and the benchmarks bullet warning against benchmarking the demo.

The status page stays gone, and the uptime worker stays undeployed. It
watched the fold cloud as well as the demo, and the fold cloud is still
down — every one of api, broker, mcp and console answers 522 — so a
status page would come up mostly red and argue against the demo the
rest of this commit is restoring. The two sentences that sent readers
to /status for live state are dropped rather than left dangling; the
page comes back when there is a cloud for it to report on.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
@blakebauman
blakebauman merged commit 98c8e3b into main Aug 26, 2026
1 check passed
blakebauman added a commit that referenced this pull request Aug 26, 2026
enterprise.fold.run has been deployed and healthy since the redeploy in
#45, and nothing anywhere pointed at it — a finished gateway with tenants,
budgets, deny-by-default policy and audit configured, that no reader could
find. This links it beside the demo, where the argument lands: it is the
same binary over a byte-identical upstream set, with the governance on.

  homepage  a governed block closing the #demo section, with the console
            and /mcp in a route rail marked "sign-in required"
  footer    "Governed gateway" in the Product column
  docs      a "same federation, governed" section in the walkthrough

Every claim is checked against the running config rather than written from
memory: both gateways federate the identical three upstreams (same ids,
urls and namespaces), auth.mode is required, policy is deny-by-default,
acme gets 5,000 calls/day at 120 req/min, globex 1,000 at 60 and no GitMCP
at all. Both curl examples in the docs were run verbatim and produce the
documented output — the 401 carries the RFC 9728 challenge, and the
metadata names oauth.work as the issuer.

It is a separate deployment because auth.mode is gateway-wide: governing
the demo would take away the copy-pasteable thing the demo exists to be.
The new copy says so rather than leaving two gateways unexplained.

Deliberately not referenced: /demo-token, which is 503 because deleting
the Worker in the teardown destroyed its four oauth.work
client-credentials secrets. Nothing shipped here depends on it, so this
points only at paths that answer today — the console (which reaches a
real login flow; the SPA client is registered) and the 401 discovery
path. Restoring those secrets is what would let the walkthrough add a
copy-paste token and go behind the wall.

Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant