Link review file paths to GitHub files#9
Conversation
|
@covencat @covencat[bot] please review this PR and confirm whether the review file references render as direct GitHub links. |
Cody dogfood resultStatus: success Files inspected Files inspected
Supporting context used
FindingsNone No-findings justification
The existing tests already assert this behavior:
Given the inspected implementation and coverage, I did not find evidence that review file references fail to render as direct GitHub links. Tests/commands considered
Confidence/limitationsConfidence is high for the specific question asked because the link-generation path and corresponding tests were directly inspected. Limitation: I did not execute the test suite in this review pass, so this conclusion is based on source and test inspection rather than a fresh runtime verification. Evidence
Structured review
Files changed: 0 Task |
|
@covencat @covencat[bot] please run one more review on this PR and specifically confirm whether the published review output turns file references like \src/adapter.ts\ into direct GitHub file links. |
Cody dogfood resultStatus: success Files inspected Files inspected
Supporting context used
FindingsNone. No-findings justificationThe published review output does turn file references like Tests/commands considered
Confidence/limitationsHigh confidence for the specific question about published review file-link conversion, because the relevant publishing path and its tests were inspected directly. Limitation: I did not execute the test suite in this review pass, so this conclusion is evidence-backed by code and test inspection rather than fresh runtime verification. Evidence
Structured review
Files changed: 0 Task |
|
@covencat @covencat[bot] please review this PR again after the hosted webhook deploy and confirm whether published file references like |
Cody dogfood resultStatus: success Files inspected Files inspectedSupporting context used
FindingsNone No-findings justificationI did not find evidence that this PR still emits non-direct file references for published paths like
Given the bounded evidence available in the changed/supporting files I inspected, I can’t substantiate a bug here. Tests/commands considered
Confidence/limitationsModerate confidence in the no-findings result, but limited by the fact that the supplied workspace did not include an explicit Evidence
Structured review
Files changed: 0 Task |
|
@BunsDev this is ready for review. Related deployment-bundle PR: OpenCoven/coven-github#70. Live post-deploy Covencat verification succeeded here: #9 (comment) |
cd18c06 to
07519a5
Compare
|
@covencat @covencat[bot] please review again after the test-command link formatter deploy. Specifically verify that Tests reported by runtime entries like �Read src/adapter.ts� link the file path portion to GitHub. |
Cody dogfood resultStatus: success Files inspected Files inspectedSupporting context used
FindingsNone No-findings justificationThe changed review-output formatter in Tests/commands considered
Confidence/limitationsHigh confidence for the specific question asked. Limitation: I reviewed the formatter implementation and existing tests only; I did not execute the test suite because no code change was requested and the review was bounded to verifying the deployed formatter behavior from source. Evidence
Structured review
Files changed: 0 Task |
07519a5 to
1d7ebac
Compare
|
@covencat @covencat[bot] please review again after the broader inline-code formatter deploy. Specifically check that narrative Tests/commands considered entries like |
Cody dogfood resultStatus: success Files inspected Files inspectedSupporting context used
FindingsNone. No-findings justificationThe relevant rendering path already links the file portion for both the narrative tests/commands entries and structured runtime tests:
Tests/commands considered
Confidence/limitationsHigh confidence for the specific requested behavior because the code path and existing assertions directly cover it. Limitation: I did not execute the test suite in this review-only run, so this assessment is based on source inspection and existing test expectations rather than a fresh runtime verification. Evidence
Structured review
Files changed: 0 Task |
1d7ebac to
2724a80
Compare
|
@covencat @covencat[bot] final verification after line-range deploy: please review whether narrative file refs like |
Cody dogfood resultStatus: success Files inspected Files inspectedSupporting context used
FindingsNone No-findings justificationThe relevant rendering path already converts both narrative refs and structured review entries into GitHub links when they are repo-relative paths:
Given that the requested examples are repo-relative file references and line ranges, the code path already supports linking them to GitHub file views. I did not find a gap in the inspected files that would prevent those references from being linked. Tests/commands considered
Confidence/limitationsConfidence is moderate-high that the reported behavior is correct based on the implementation inspected. Limitation: I did not execute the test suite or inspect the full PR patch diff beyond the supplied comment context, so this review is bounded to the relevant rendering and publication code paths visible in Evidence
Structured review
Files changed: 0 Task |
2724a80 to
1ddcfd4
Compare
|
@covencat @covencat[bot] final check after bracket-identifier guard: verify |
Cody dogfood resultStatus: success Files inspected Files inspectedSupporting context used
FindingsNone No-findings justificationThe relevant linkification code in Tests/commands considered
Confidence/limitationsHigh confidence that the reported bracket-identifier guard is functioning as described for Evidence
Structured review
Files changed: 0 Task |
1ddcfd4 to
b1fd039
Compare
|
@covencat @covencat[bot] verify the missing bare-path case from the screenshot: in Tests/commands considered, entries like Read src/adapter.ts, Read tests/webhook-adapter.test.ts, and Read AGENTS.md should link those bare file paths to GitHub, while existing URLs and tests_run[].output_summary should not be rewritten as file links. |
Cody dogfood resultStatus: success Files inspected Files inspectedSupporting context used
FindingsNone. No-findings justificationThe reviewed code already matches the requested behavior described in the issue comment. In
The corresponding tests in
That directly covers the screenshot-style bare-path case and the non-rewrite requirement for Tests/commands considered
Confidence/limitationsConfidence is high for the specific behavior mentioned in the issue comment because the implementation and tests already cover it directly. Limitation: I only inspected the targeted files relevant to the comment, per the bounded review instructions, so this review does not make claims about unrelated parts of the repo. Evidence
Structured review
Files changed: 0 Task |
b1fd039 to
2fcb731
Compare
|
@BunsDev ready for review. |
|
@BunsDev updated with the matching trigger-gating fix: plain \issues.opened\ deliveries are ignored, and routing now respects webhook-style \enabled_triggers\ values before queueing work. Validated with |
|
@covencat please perform a fresh, real code review of current head d60fe68 against the current main branch. Be deliberately critical and review every changed file. Check correctness, security, regressions, merge compatibility with current main, and missing or misleading tests. Run relevant tests where the environment permits. If actionable issues exist, list each one with severity, exact file and line, why it matters, and the specific change you recommend. Make the result suitable for a requested-changes decision. Do not modify the branch. Do not invent findings. Report no findings only after reconciling the full changed-file set and test evidence. |
![covencat[bot]](https://avatars.githubusercontent.com/in/4224630?s=60&v=4){kind=small}
There was a problem hiding this comment.
Coven review
Status: failure
The runtime review output was downgraded because its publication evidence was incomplete or contradictory.
Evidence
- PR: #9
- Base:
main@52c6d2bb9ba46f76c5eb1692cff1e9c9b744ae56 - Head:
agent/github-review-file-links@d60fe686dbb02f0ed07de99ae36fd39197711812 - Checked-out workspace HEAD:
d60fe686dbb02f0ed07de99ae36fd39197711812 - Changed files supplied to agent: 2
- Review context SHA-256:
df700d6176937c3baece100facf952d152faeee5651a18f82d4d2a984d655ed9 - Files:
src/adapter.ts,tests/webhook-adapter.test.ts
Structured review
- Mode:
pull_request - Evidence status:
partial - Reviewed files: 2
- Reviewed file list:
src/adapter.ts,tests/webhook-adapter.test.ts - Supporting files inspected: 0
- Findings: 0
- Tests reported by runtime: 0
- Limitations: 3
- No supporting-code file reads or search results were captured during this review.
- Review narrative was generic and did not explain the review outcome.
- Review output did not include structured findings, a file-backed no-findings justification, or an explicit limitation explaining why substantive review was not possible.
Files changed: 0
Commits: 1
Task 0a39bf30-7fb9-11f1-986a-b63c4a70fa13.
Publication validation
- review evidence was not marked complete
- runtime result was not successful
- captured diff evidence is incomplete or truncated
- review reported limitations
- no test execution evidence was reported
- no-findings review is missing its justification
Evidence was incomplete or contradictory, so this is a COMMENT review rather than an approval or change request.
There was a problem hiding this comment.
Coven review
Status: success
Files inspected
Files inspected
src/adapter.tstests/webhook-adapter.test.ts
Supporting context used
AGENTS.md— repository contribution/review constraints and workflow guidance.src/adapter.tssections around review-context handling, evidence generation, scope validation, and finalize/publish flow (prepareReviewContextcall site,reviewEvidence,expectedReviewScopeStatement,trustedValidationFindings,finalizeReviewResult).tests/webhook-adapter.test.tssections covering the new brief/instruction expectations, review-task fixture shape, trusted-validation mapping, and review-finalization behavior.
Findings
None
No-findings justification
The inspected change set is internally consistent with the supporting review/policy code. In src/adapter.ts, the review path now attaches trusted validation receipts to review_context, persists a SHA-256, and records evidence before running the review cycle, which matches the downstream consumers in reviewEvidence and finalizeReviewResult (src/adapter.ts:1852-1870, src/adapter.ts:3063-3090, src/adapter.ts:2767-2795). The scope/language checks also align with the new brief instructions: expectedReviewScopeStatement accepts bounded-scope statements while rejecting material limitations, and the test suite explicitly exercises that behavior (src/adapter.ts:2717-2723, tests/webhook-adapter.test.ts:1130-1163, tests/webhook-adapter.test.ts:1165-1187). The trusted-validation derivation and finalization tests likewise cover the changed-file mapping and limitation filtering behavior (tests/webhook-adapter.test.ts:2156-2196).
Tests/commands considered
git diff --check- passed: no whitespace or patch-format issues in the reviewed commit.
Confidence/limitations
None
Evidence
- PR: #9
- Base:
main@6a82b4378ab531b5d38475e1fd2f7ddbdd74632f - Merge base:
6a82b4378ab531b5d38475e1fd2f7ddbdd74632f - Head:
agent/github-review-file-links@7ba243120ebbab9f1fd12fb1d29de22949d7740f - Checked-out workspace HEAD:
7ba243120ebbab9f1fd12fb1d29de22949d7740f - Changed files supplied to agent: 2
- Review context SHA-256:
c6da204e1fef68e44354c7b0b30e137134d02f04c954267910407cbdd65b6e23 - Files:
src/adapter.ts,tests/webhook-adapter.test.ts
Structured review
- Mode:
pull_request - Evidence status:
complete - Reviewed files: 2
- Reviewed file list:
src/adapter.ts,tests/webhook-adapter.test.ts - Supporting files inspected: 1
- Supporting file list:
AGENTS.md - Findings: 0
- No-findings reason: The inspected change set is internally consistent with the supporting review/policy code. In
src/adapter.ts, the review path now attaches trusted validation receipts toreview_context, persists a SHA-256, and records evidence before running the review cycle, which matches the downstream consumers inreviewEvidenceandfinalizeReviewResult(src/adapter.ts:1852-1870,src/adapter.ts:3063-3090,src/adapter.ts:2767-2795). The scope/language checks also align with the new brief instructions:expectedReviewScopeStatementaccepts bounded-scope statements while rejecting material limitations, and the test suite explicitly exercises that behavior (src/adapter.ts:2717-2723,tests/webhook-adapter.test.ts:1130-1163,tests/webhook-adapter.test.ts:1165-1187). The trusted-validation derivation and finalization tests likewise cover the changed-file mapping and limitation filtering behavior (tests/webhook-adapter.test.ts:2156-2196). - Tests reported by runtime: 1
git diff --check:passed- Command completed successfully with no output.
- Limitations: 0
Files changed: 0
Commits: 1
Task autoreview-ea9f271099a682354174ee10f6cd1650dd4ea749ddbca381.
This review follows the prior covencat publication. A decisive submission replaces its state; a COMMENT does not.
1 participant
Summary
Make Covencat's published review output turn safe repository-relative file references into direct GitHub blob links pinned to the correct reviewed revision.
This applies to summary and PR-body prose, structured reviewed/supporting file lists, findings, and test-command evidence such as
Read src/adapter.ts.Review resolution
This head incorporates current
mainand resolves the issues found during a full review:mainremains intact.Validation
npm test— 123/123 passed.npm run build— passed.node scripts/demo-app-smoke.mjs— passed.git diff --check— passed.TypeScript test + build— passed on head7ba2431.Scope
Only
src/adapter.tsandtests/webhook-adapter.test.tsdiffer from currentmain.Related work
Those comments exercised earlier iterations. This reviewed head is verified by local and GitHub CI; this PR does not claim that the new head has been deployed to production.