============
Slice Review
============

Lightweight two-axis review of the diff since a fixed point — Standards
and Spec — run as independent personas.

Description
-----------

The slice-review skill is the per-slice quality gate used by
``implement`` (and available standalone). It reviews
``git diff <fixed-point>...HEAD`` along two deliberately separate axes:

- **Standards** — conformance to the repository's documented coding
  standards (``AGENTS.md``, ``CLAUDE.md``, ``CONTRIBUTING.md``) plus a
  fixed baseline of Fowler code smells, applied as judgement calls that
  documented repo standards always override.
- **Spec** — fidelity to the originating slice/spec under
  ``resources/agent-docs/specs/``: missing requirements, scope creep, and
  wrong-looking implementations, each with the spec line quoted.

The two personas run independently so neither pollutes the other's
context, and their findings are reported side by side, never merged or
reranked — code can pass one axis and fail the other.

For whole-spec review, use ``team-review-impl`` instead.

Triggers
--------

.. code-block:: text

   /wf:slice-review master
   /wf:slice-review HEAD~3

Also: "review this slice", "review since <ref>".

Output
------

A report with ``## Standards`` and ``## Spec`` sections and a one-line
summary of finding counts and the worst issue per axis.

Attribution
-----------

Adapted from the ``code-review`` skill in
`mattpocock/skills <https://github.com/mattpocock/skills>`_ (MIT
license). Renamed to mark its per-slice role.