diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index 6f1f41c..7857912 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -16,3 +16,11 @@ jobs:
           version: ${{ matrix.julia-version }}
       - uses: julia-actions/julia-buildpkg@v1
       - uses: julia-actions/julia-runtest@v1
+      - uses: julia-actions/julia-processcoverage@v1
+      - uses: codecov/codecov-action@v5
+        with:
+          files: lcov.info
+          token: ${{ secrets.CODECOV_TOKEN }}
+          # The org-wide token cannot imply a repository; the slug says
+          # which repo this upload is for.
+          slug: ${{ github.repository }}
diff --git a/README.md b/README.md
index 23db01b..1136239 100644
--- a/README.md
+++ b/README.md
@@ -1,6 +1,6 @@
 # ExaModelsAMPL.jl
 
-Write an [ExaModels.jl](https://github.com/exanauts/ExaModels.jl) model to AMPL's
+Write an [ExaModels.jl](https://github.com/madsuite-org/ExaModels.jl) model to AMPL's
 `.nl` interchange format, so it can be consumed by any ASL-based tool (AMPL-compatible
 solvers, [AmplNLReader.jl](https://github.com/JuliaSmoothOptimizers/AmplNLReader.jl), ...)
 without building a JuMP model first.
diff --git a/codecov.yml b/codecov.yml
new file mode 100644
index 0000000..26b9d44
--- /dev/null
+++ b/codecov.yml
@@ -0,0 +1,10 @@
+# Coverage numbers without CI redness: the project/patch checks report their
+# figures but never fail — a coverage drop is information, not a broken build.
+coverage:
+  status:
+    project:
+      default:
+        informational: true
+    patch:
+      default:
+        informational: true