test: prove ~520-char stateless GitHub installation tokens round-trip - #537
Merged
seonghobae merged 4 commits intoSep 4, 2026
Merged
Conversation
…trips
GitHub announced that GitHub App installation tokens are moving to a new
stateless ghs_... format that can be roughly 520 characters, up from the
historical ~40. Audited createInstallationToken() and confirmed
githubInstallationTokenPattern (src/index.ts) already validates with
/^[\x21-\x7e]{1,4096}$/ -- a printable-ASCII, 1-4096-length pattern with
no fixed-length assumption -- so no source change is required. This adds
a regression test locking in that a ~520-character token exchanges
successfully end-to-end, since no existing test exercised a token longer
than the historical format.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
|
Warning Review limit reachedNext included review available in 57 minutes. View limit detailsLimit details: You’ve used the included review currently available. You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository. Review configuration: ⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Team Run ID: 📒 Files selected for processing (1)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
seonghobae
added
priority: medium
ChatGPT Codex Connector
This was referenced Sep 3, 2026
seonghobae
marked this pull request as draft
September 3, 2026 14:37
seonghobae
marked this pull request as ready for review
September 4, 2026 08:39
seonghobae
deleted the
claude/installation-token-stateless-format-regression
branch
September 4, 2026 08:40
seonghobae
added a commit
that referenced
this pull request
Sep 4, 2026
seonghobae
added a commit
that referenced
this pull request
Sep 4, 2026
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
1 participant
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.
Scope
Prove that Noema's GitHub App installation-token broker remains compatible with GitHub's 2026 stateless installation-token rollout without adding token introspection or a second format authority.
GitHub's primary rollout guidance says new installation tokens retain the
ghs_prefix, are roughly 520 characters, use a stateless dot-bearing transport shape, and must be treated as opaque by clients. Production already accepts printable-ASCII installation tokens up to 4096 characters and has no fixed short-token assumption, so no production change is justified. The regression drives a representative ~520-character stateless token through the full mocked/exchangepath and asserts byte-for-byte round-trip; Noema never decodes or semantically validates those opaque bytes.The predecessor fixture had only
ghs_plus repeatedAcharacters. Commit009187d6e70e5aec2f2c47630a87590cc98bcb55repaired that evidence defect by exercising the documented dot-bearing shape without changing production behavior.Protected-base convergence
After #528 reached protected truth, the branch was non-force merged with fresh protected
main@bbee33270b496255d785c766fc009a5f9162a695, preserving its single test-file delta. Current exact head is81222c10f5b018a07b4e7df81dcd589b28285a13; no predecessor check or review evidence transfers.Current exact authority — 2026-09-04 KST
main@bbee33270b496255d785c766fc009a5f9162a695;81222c10f5b018a07b4e7df81dcd589b28285a13;ci 33698029634: terminal success on this head;Security Scan 33698029622: terminal success on this head;patch-validator-image 33698029606/ job100471061188: terminal success on this exact head. The runner checked out the exact head and completed static runtime identity, exact image metadata, real no-network/read-only/non-root smoke, CycloneDX SBOM, image/static-runtime/embedded-runtime vulnerability receipts, exact-source/image receipt verification, stale-head refusal and bounded evidence upload;reviewer-ci 33698029578: terminal workflow success but not valid semantic GREEN. Job100471060997indexed its CodeGraph fixture and returnedNo relevant code found ...; the old gate accepted wrapper banners, so this is false-green evidence addressed by fix(reviewer): fail closed on empty CodeGraph semantics #546.The image/SBOM/vulnerability/provenance lane is complete, but reviewer authority is not. Fresh #546 authority is exact
fb3510e8f5e5613c8b5a0bf9cebb3a6395688fe1, Draft/mergeable. In addition to rejecting no-relevant/empty/final-section-spoof cases, its RED58d5876313e55aee6a0e976ba4bff1ca2fa0ce4bproves a synthetic[truncated N characters]annotation with no retained semantic bytes cannot count as CodeGraph evidence; productionfb3510e8...fails closed on that annotation-only final explore section. Exact #546 runsci 33790683473,reviewer-ci 33790683468, requiredSecurity Scan 33790683516, andpatch-validator-image 33790683437remain queued before executable verification and therefore non-passing. This PR must remain Draft/non-merge-ready until #546 reaches protected truth and this unchanged head acquires fresh reviewer evidence under the repaired gate.Do not transfer the old reviewer success, mutate this clean head merely to retrigger scheduling, self-approve, force-push/destructively rebase, or weaken required gates.