Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 13 additions & 0 deletions review-suite/evals/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -40,6 +40,11 @@ review-suite/evals/
│ ├── SOURCING.md ground truth, sanitization, batches
│ ├── LIMITATIONS.md explicit inputs to interpretation
│ └── pilot/<stratum>.report.json the unscored pilot's compact reports
├── v2/ the v2 planning/definition record (#59)
│ ├── FAILURE-TAXONOMY.md every v1 material outcome, classified
│ ├── DECISION-RECORD.md per-mechanism evidence and disposition
│ ├── gate-manifest.json preregistered v2 gates, before scoring
│ └── audits/ the three graph audits (later step, empty for now)
└── artifacts/ opt-in captured output, not in git

review-suite/scripts/evals/
Expand Down Expand Up @@ -70,6 +75,14 @@ figure. In particular: **the connector stratum is deferred, not satisfied.**
Connector-escape recall has never been measured here, and no human-review figure
may be reported as a connector figure.

## v2 planning

[`v2/README.md`](v2/README.md) records the planning/definition gate that reads
this frozen v1 baseline and turns it into an implementation-ready decision for
the review suite's next mechanisms, before any scored v2 output exists. It
changes no review prompt, schema, rubric, orchestration, or caller runtime
behavior by itself.

## Calibration

An uncalibrated grader does not report a conservative score; it reports a
Expand Down
Loading
Loading