Objective
Retire the top-level \ CLI surface and move rubric-generation guidance/workflow into skill-based eval authoring, preferably by expanding \ rather than introducing another narrowly-scoped skill. The current \ namespace only fronts , while eval authoring already lives in the skill/docs layer, which leaves AgentV with overlapping ways to do the same authoring task.
Architecture Boundary
Primary boundary: external-first eval-authoring workflow and docs. Keep core/runtime changes minimal and focused on removing a redundant CLI entrypoint plus any deprecation or compatibility shim needed for migration.
Design Latitude
- Preferred direction: teach rubric generation through \ and related docs, then remove .
- Acceptable alternative: if the authoring flow does not fit cleanly inside , introduce a dedicated skill, but only if that is materially cleaner than extending the existing eval-writer skill.
- The migration can use a staged deprecation if immediate removal would create too much churn.
- Keep the user-facing authoring story coherent across CLI help, docs, examples, and skill guidance.
Acceptance Signals
- \ is removed or clearly deprecated with a migration path to the supported skill-based authoring workflow.
- \ (or a justified replacement skill) covers rubric-generation guidance well enough that users do not need the CLI subcommand for this task.
- Docs stop presenting \ as the default path and instead point users to the supported authoring workflow.
- Error/help text that currently points users at \ is updated to reference the new workflow.
- Tests/docs/examples are updated so there is one clear supported path for rubric authoring.
Non-Goals
- Reworking the rubric evaluator itself.
- Redesigning the full eval file schema.
- Expanding scope into broader skill marketplace or trigger-quality work.
Context
- \ currently exposes only a single namespace: .
- Docs currently duplicate this path in CLI/web docs, while eval authoring guidance already exists in \ and the skill-improvement docs.
- This makes rubric generation feel like core product surface area even though it is closer to assisted eval authoring.
Related
Objective
Retire the top-level \ CLI surface and move rubric-generation guidance/workflow into skill-based eval authoring, preferably by expanding \ rather than introducing another narrowly-scoped skill. The current \ namespace only fronts , while eval authoring already lives in the skill/docs layer, which leaves AgentV with overlapping ways to do the same authoring task.
Architecture Boundary
Primary boundary: external-first eval-authoring workflow and docs. Keep core/runtime changes minimal and focused on removing a redundant CLI entrypoint plus any deprecation or compatibility shim needed for migration.
Design Latitude
Acceptance Signals
Non-Goals
Context
Related