@christianmorup/review-intent 0.1.0 → 0.1.3

package/README.md

@@ -92,10 +92,12 @@ self-contained inline SVG (no charting dependency, deterministic output):
    that carry agent rationale (the completeness contract, visualized).
 4. **Reach ripple** — the reach graph as concentric rings: changed files at the
    centre, importers rippling outward.
-5. **Honesty quadrant** — the signature view: measured *blast radius* (churn +
-   reach) on the x-axis against claimed *candor* (intent coverage + declared
-   risks) on the y-axis. A dot landing in the shaded red corner is a high-impact
-   change that declared little risk — the contradiction made into a picture.
+5. **Change map** — the signature view: one dot per changed file, placed by
+   measured *downstream reach* (how many repo files import it) on the x-axis
+   against measured *churn* (± lines) on the y-axis, sized by hunk count and
+   coloured red when the file carries a complexity hotspot. Files in the shaded
+   top-right corner are the largest and most depended-on — the review-first
+   targets, picked out instead of being averaged into a single PR-wide score.
 
 `npm run sample` builds and writes a representative `sample-output.html` you can
 open to see the whole page.
@@ -206,7 +208,7 @@ whether the intent is genuine reasoning or post-hoc rationalization is the whole
 ballgame. `review-intent` ships a Claude Code skill that teaches the agent to
 author the artifact **honestly** (real rejected alternatives, stated
 assumptions, incidental changes marked as incidental) when it finishes a change
-set, then offer to render it.
+set, then render it.
 
 ```sh
 review-intent skill install            # ~/.claude/skills/review-intent-authoring (all repos)
@@ -215,9 +217,10 @@ review-intent skill uninstall          # remove user-scoped skill
 review-intent skill uninstall --local  # remove repo-scoped skill
 ```
 
-The skill never auto-launches anything — it teaches the agent to write the
-artifact and then *ask* before opening the review. Add `--force` to overwrite or
-remove a hand-edited skill file. User and `--local` scopes are independent.
+Once the artifact is written the skill tells the agent to *run* `review-intent`
+and open the review — not to ask first, since invoking the skill already means a
+review is wanted. Add `--force` to overwrite or remove a hand-edited skill file.
+User and `--local` scopes are independent.
 
 The honesty contract is the point: a fluent rationalization is worse than
 nothing because it lowers the reviewer's guard while adding no signal. The skill