Skip to content

Package complete release evidence - #83

Merged
Anionix merged 2 commits into
mainfrom
codex/issue-73-release-completeness
Jul 14, 2026
Merged

Package complete release evidence#83
Anionix merged 2 commits into
mainfrom
codex/issue-73-release-completeness

Conversation

@Anionix

@Anionix Anionix commented Jul 14, 2026

Copy link
Copy Markdown
Owner

Fixes #73

  • package every referenced and generated artifact under safe relative paths
  • reject missing or escaping artifact references
  • include normalized result hashes in fair Markdown
  • include schema bytes in prompt Markdown
  • make README commands runnable through the frozen uv environment

Verification: 55 tests passed. Release archives and generated reports will be regenerated from the final public commit.

@Anionix

Anionix commented Jul 14, 2026

Copy link
Copy Markdown
Owner Author

@codex review

@chatgpt-codex-connector chatgpt-codex-connector Bot 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.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: fa2d357394

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment thread src/format_bench/release.py Outdated
Comment on lines +34 to +35
if key == "artifact" and isinstance(value, str):
references.append(value)

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

P1 Badge Restrict artifact discovery to run artifact fields

For claims runs, negative_research is loaded from research/formats/*.json; research/formats/anyblox.json has source_commits.artifact set to a commit SHA rather than a run file. Because this recursive scan treats every nested key named artifact as a run-relative path, packaging a claims run will look for <run_dir>/3cf5008b4af5165761021b9f6f24d326d83d2a56 and fail before the benchmark workflow can upload the release archive. Limit discovery to the known manifest format entries and result artifact maps instead of arbitrary nested metadata keys.

Useful? React with 👍 / 👎.

Copy link
Copy Markdown
Owner Author

Choose a reason for hiding this comment

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

Fixed in 2d05381. Artifact discovery now reads only manifest format entries and the formal artifact/artifacts fields on benchmark observations (or their evidence object). A regression test includes negative_research.source_commits.artifact and confirms it is not treated as a run path.

@Anionix

Anionix commented Jul 14, 2026

Copy link
Copy Markdown
Owner Author

@codex review

@chatgpt-codex-connector chatgpt-codex-connector Bot 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.

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 2d0538133e

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

Comment thread src/format_bench/release.py
@Anionix
Anionix merged commit d89ac48 into main Jul 14, 2026
4 checks passed
@Anionix
Anionix deleted the codex/issue-73-release-completeness branch July 14, 2026 04:08
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.

Complete release packages, fair hashes, and runnable README commands

1 participant