Cut 0.4.0 - #40
Merged
Merged
Conversation
Minor rather than patch. 0.3.0 predates the evidence report entirely, so anyone who installed from a package rather than source has no way to export what they were shown — the project promises evidence and today it dies when the tab closes. New public surface: EvidenceReport, AnalysisResult.Signals and .Observations, Finding.AtHumanRate, ReportOptions.InterfaceLanguage, and write_report as the tenth MCP tool. The calibration was regenerated before this build rather than after, so the embedded snapshot names the engine that carries it. Running it the other way round ships a package whose printed error rate was measured by a different version — the numbers are unchanged from yesterday, which is itself the result worth having: the genre gating did not move the published rate. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Version bump across the three packages and
server.json(twice), plus the regenerated calibration snapshot.Minor, not patch. 0.3.0 predates the evidence report entirely: anyone who installed from a package rather than from source cannot export what they were shown. The project promises evidence, and until this release that evidence dies when the tab closes.
New public surface:
EvidenceReport,AnalysisResult.Signals/.Observations,Finding.AtHumanRate,ReportOptions.InterfaceLanguage, andwrite_reportas the tenth MCP tool.Order matters here and it is easy to get wrong. The calibration tool has a
ProjectReferenceto Core, so Core compiles before the run writes the snapshot — and the snapshot records the engine version it was measured by. Versions were bumped first, then the calibration regenerated, then Core rebuilt, and the embedded resource checked against the source file so the shipped assembly cannot disagree withDocs/CALIBRATION.md. Tracked as part of #37.The numbers themselves are unchanged from yesterday, which is the result worth having: the genre gating in #34 did not move the published false-positive rate.
333 tests green.
🤖 Generated with Claude Code