docs(readme): state that this is one adapter, not the way to consume workspace.json - #14
Merged
Merged
Conversation
![greptile-apps[bot]](https://avatars.githubusercontent.com/in/867647?s=60&v=4){kind=small}
There was a problem hiding this comment.
qmarcelle has reached the 50-credit limit for trial accounts. To continue receiving code reviews, upgrade your plan.
Reviewer's guide (collapsed on small PRs)Reviewer's GuideClarifies README to position workspace.json as a host-neutral standard and this repository as the Codex adapter, and explains the Billfold demo link’s differing namespace as intentional. File-Level Changes
Tips and commandsInteracting with Sourcery
Customizing Your ExperienceAccess your dashboard to:
Getting Help
|
Contributor
Author
|
@sourcery-ai review |
![sourcery-ai[bot]](https://avatars.githubusercontent.com/in/48477?s=60&v=4){kind=small}
qmarcelle
force-pushed
the
gtm-39-host-neutral-framing
branch
from
August 13, 2026 00:07
0ee76bc to
c3c17a2
Compare
There was a problem hiding this comment.
qmarcelle has reached the 50-credit limit for trial accounts. To continue receiving code reviews, upgrade your plan.
…workspace.json A reader landing here met a Codex lockup, a Codex-specific positioning line, a Codex package name, and then every operational section in Codex terms. Nothing said the standard is host-neutral, so the available reading was that workspace.json is a Codex feature. That inverts the topology the standard actually defines, where the committed artifact is the interoperability point and no integration is privileged. Adds a "Where this fits" section before the Codex material: the format is host-neutral, this repository holds host adapters, Codex is the adapter that exists today, and further adapters can be added without the format changing. The ownership table names which repository owns the format, the generation and the adapters, so the reader can place all three without leaving the page. Deliberately not overstated. This repository currently *is* the Codex adapter — its root package is `@workspacejson/codex-mcp` — so the text says Codex is the one that exists rather than implying a roster that is not there. The lockup and the Codex-specific sections stay where they are, because they accurately describe what this package does; the positioning line now marks it as the Codex adapter rather than as the subject of the standard. Also notes that `billfold` is hosted under the superseded org, so the differing URL reads as a known state rather than a mistake. That link is correct and was left alone: the repository has not been transferred, and the canonical-org URL does not resolve.
qmarcelle
force-pushed
the
gtm-39-host-neutral-framing
branch
from
August 13, 2026 00:10
c3c17a2 to
3aa4531
Compare
There was a problem hiding this comment.
qmarcelle has reached the 50-credit limit for trial accounts. To continue receiving code reviews, upgrade your plan.
There was a problem hiding this comment.
qmarcelle has reached the 50-credit limit for trial accounts. To continue receiving code reviews, upgrade your plan.
|
qmarcelle
added a commit
that referenced
this pull request
Aug 13, 2026
…fy (GTM-45) (#17) `Greptile Review` was promoted to a required context on 2026-08-12 after a seven-criterion calibration. Criterion 5 — a further push retriggers review against the new head — stopped holding the next day. Greptile kept posting reviews and stopped emitting the check run branch protection matches on, so the required context could not be satisfied by any action available to a contributor and `main` became unmergeable for every change, hotfixes included. Observed on PR #14 across four heads (`0ee76bc`, `c3c17a2`, `3aa4531`, `be2e965`): a Greptile review posted on each, zero `Greptile Review` check runs on any, 8/8 other checks green, zero unresolved threads. The same `statusCheck: true` kept producing the check on `workspacejson/standard` across #34, #35 and #36 in the same window — the configuration is not missing, the emission is not dependable. Applied to protection on 2026-08-13, measured before and after against the API rather than trusting the write response. App id bindings preserved on the four surviving contexts. Greptile is **not** uninstalled — `greptile-apps` (867647) remains installed and reviewing; only its authority over merge eligibility is withdrawn. Criterion 5 is annotated rather than deleted: it is the criterion whose failure the requirement could not survive, and the calibration record now says what the protocol did not test — durability. Every criterion was measured once, in one sitting. A behavioral bar measured once is a measurement, not a guarantee. Recorded honestly: PR #14 merged by administrator bypass, not by satisfying protection, so the GTM-39 receipt rests on four contexts plus conversation resolution. A gate satisfiable only by bypass is not a gate. The accepted cost is that `main` now has no mechanically enforced reviewer context at all — conversation resolution and the written per-finding protocol carry the whole semantic gate. Re-admission requires both a substantive review on the current head and a mechanically enforceable current-head signal. Absence is reported as absence; quota and error comments are not review evidence.
qmarcelle
added a commit
that referenced
this pull request
Aug 13, 2026
…TM-45) (#18) The withdrawal record said "emission on this repository is not dependable." That was accurate about the symptom and wrong about the cause, and the difference decides what anyone does next: a flaky app is waited out, an exhausted trial is topped up. Established 2026-08-13. Greptile posts, in place of a review: `qmarcelle` has reached the 50-credit limit for trial accounts. To continue receiving code reviews, upgrade your plan. and emits no check run. That is why the check fired on #10 and #13 and then stopped at #14 — the credits ran out in between, not the app's emission becoming unreliable here. The prediction that follows was confirmed the same day. `workspacejson/standard` kept working only until its own next pull request: #37 reviewed the exact head 4f9e8f6f and emitted zero check runs, against exactly one each on #34, #35 and #36. It hit the identical deadlock and its requirement was withdrawn the same way. So the "next door still works" observation in the original evidence was true but temporary, and reading it as repository-specific would have sent the investigation somewhere there was nothing to find. Restoring the signal is a billing action.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
A reader landing here met a Codex lockup, a Codex-specific positioning line, a
Codex package name, and then every operational section in Codex terms. Nothing
on the page said the standard is host-neutral, so the available reading was that
workspace.jsonis a Codex feature.That inverts the topology the standard defines: the committed artifact is the
interoperability point, and no integration is privileged. It is the same class
of error as a category claim outrunning its artifact.
What this adds
A "Where this fits" section ahead of the Codex material:
What it deliberately does not do
It does not pretend this is a multi-adapter monorepo. The root package here
is
@workspacejson/codex-mcp. So the text says Codex is the one that exists,rather than implying a roster that is not there — an overcorrection would just be
a different inaccuracy.
The lockup and the Codex-specific sections stay where they are. They accurately
describe what this package does. Only the positioning line changes, to mark it
as the Codex adapter rather than as the subject of the standard.
No asset was redesigned. If a neutral integrations lockup is wanted, that is a
separate request against the design kit.
One link note
billfoldnow carries a sentence saying it is hosted under the supersededworkspace-jsonorg, so the differing URL reads as a known state rather than amistake.
The link itself was left alone deliberately: that repository has not been
transferred, and
workspacejson/billfoldreturns 404. Rewriting it to thecanonical namespace would hand readers a broken clone command. Every GitHub link
in the README was checked against the API; all four resolve.
lintandcheck:structurepass.Summary by Sourcery
Clarify the role of this repository as the Codex adapter for the host‑neutral workspace.json standard and position Codex as one integration rather than the standard itself.
Documentation: