Skip to content

docs: define checkout-native review contract#377

Merged
rianjs merged 1 commit into
mainfrom
issue-367-checkout-native-review-contract
Jun 24, 2026
Merged

docs: define checkout-native review contract#377
rianjs merged 1 commit into
mainfrom
issue-367-checkout-native-review-contract

Conversation

@rianjs

@rianjs rianjs commented Jun 24, 2026

Copy link
Copy Markdown
Contributor

No description provided.

@rianjs

rianjs commented Jun 24, 2026

Copy link
Copy Markdown
Contributor Author

Findings

No findings.

The PR is docs-only and stays aligned with ticket #367’s role in the batch: it establishes a shared contract without prematurely changing pipeline behavior. The contract covers the architectural pressure points we discussed: no diff/full-file context stuffing, dossier/workbench sequencing, run-owned artifacts and retention, SelectionOnly/benchmark compatibility, fork-backed PR support, artifact-digest fingerprints for path-referenced prompts, adapter no-fallback behavior, and reviewer coverage reporting for safe rollup.

Residual implementation risk remains in later tickets, especially adapter enforcement and fingerprint correctness, but this contract gives those tickets a clear boundary.

@rianjs

rianjs commented Jun 24, 2026

Copy link
Copy Markdown
Contributor Author

Minor: the current suite does not really prove the new checkout-native contract.

The closest coverage in internal/pipeline/pipeline_test.go and internal/cmd/benchmarkcmd/select_test.go only asserts the legacy selection-only path, caller-owned artifact roots, and prompt provenance hashes. The task-artifact tests in internal/pipeline/pipeline_test.go and internal/pipeline/pipeline_test.go cover metadata safety and a subset of progress fields, but nothing here would fail if the follow-on implementation still stuffed full diffs, reused stale dossier/workbench artifacts, ignored allowed_files, or assumed same-repo PRs only. That is acceptable for a docs-only PR, but the implementation issues should add focused regressions for the new order, artifact-digest invalidation, and checkout-readonly adapter behavior.

@monit-reviewer monit-reviewer left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Automated PR Review

Reviewed commit: 1094a31

Summary

No issues found.

2 PR discussion threads considered.


Completed in 49s | $0.83 | sonnet | daemon 0.2.132 | Glorfindel
Field Value
Model sonnet
Reviewers hybrid-synthesis, documentation:docs-reviewer, harness-engineering:harness-architecture-reviewer, harness-engineering:harness-enforcement-reviewer, harness-engineering:harness-knowledge-reviewer
Engine claude · sonnet
Reviewed by pr-review-daemon · monit-pr-reviewer
Duration 49s wall · 46s compute (Reviewers: 35s · Synthesis: 11s)
Cost $0.83 (estimated)
Tokens 180.9k in / 4.0k out
Turns 10

Per-workstream usage

Workstream Model In Out Cache read Cache create Cost
hybrid-synthesis sonnet 33.1k 242 13.1k 19.9k (1h) $0.13
documentation:docs-reviewer sonnet 36.2k 1.0k 13.1k 23.0k (1h) $0.16
harness-engineering:harness-architecture-reviewer sonnet 36.4k 284 2.6k 33.8k (1h) $0.21
harness-engineering:harness-enforcement-reviewer sonnet 37.9k 1.4k 13.1k 24.7k (1h) $0.17
harness-engineering:harness-knowledge-reviewer sonnet 37.4k 1.0k 13.1k 24.2k (1h) $0.16

Re-reviews only run when @monit-reviewer is re-requested as a reviewer — push as many commits as you need, then re-request when ready. PRs targeting branches other than main, master are skipped, even when @monit-reviewer is re-requested.

@rianjs rianjs marked this pull request as ready for review June 24, 2026 15:41
@rianjs rianjs merged commit 4a69258 into main Jun 24, 2026
10 checks passed
@rianjs rianjs deleted the issue-367-checkout-native-review-contract branch June 24, 2026 15:41
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.

2 participants