diff --git a/.markdownlint.yml b/.markdownlint.yml
new file mode 100644
index 0000000..bf8c71d
--- /dev/null
+++ b/.markdownlint.yml
@@ -0,0 +1,6 @@
+# Allow creation of collapsable sections in Markdown
+# (See https://docs.github.com/en/get-started/writing-on-github/working-with-advanced-formatting/organizing-information-with-collapsed-sections)
+MD033:
+  allowed_elements:
+  - summary
+  - details
diff --git a/docs/decisions/0002-automate-testing-of-platform.md b/docs/decisions/0002-automate-testing-of-platform.md
new file mode 100644
index 0000000..8f7b9de
--- /dev/null
+++ b/docs/decisions/0002-automate-testing-of-platform.md
@@ -0,0 +1,85 @@
+# Approach for Automated Testing of the Platform
+
+* Status: proposed
+* Deciders: [list everyone involved in the decision] <!-- optional -->
+* Date: [YYYY-MM-DD when the decision was last updated] <!-- optional -->
+
+Technical Story: Spike: Automated testing for platform [[#6](https://github.com/navapbc/platform/issues/6)
+
+## Context and Problem Statement
+
+Testing the platform currently requires multiple time-consuming steps, involving
+creating a new repo, copying over template code, creating infrastructure from
+template code, and verifying that CI and CD works. Furthermore, splitting up the
+platform into separate template-infra and template-application-* repositories
+increases the importance of testing the pieces together in an integrated
+fashion. In order to sustainably maintain development of the platform, we need a
+way to test the platform in an automated fashion.
+
+## Proposal
+
+### Overview
+
+* Have a CI GitHub Action in `platform` repo that runs regularly (e.g. weekly or daily)
+* The `platform` CI will:
+  1. Create a test repo `platform-test-[WORKFLOW_RUN_ID]`
+  2. Install the `template-infra` into the test repo
+  3. Install `template-application-nextjs` into the test repo
+  4. Run through the steps to set up the terraform backend
+  5. Spin up the application's infrastructure
+  6. Push up a test commit
+  7. Watch the GitHub action workflow runs for CI and CD
+  8. Check that CI and CD both complete successfully
+  9. Hit the deployed application's health check endpoint
+  10. Destroy the infrastructure
+  11. Delete the test repo
+
+### Details
+
+TBD
+
+## Decision Outcome
+
+TBD
+
+### Positive Consequences
+
+TBD
+
+### Negative Consequences
+
+TBD
+
+## Pros and Cons of the Options <!-- optional -->
+
+### [option 1]
+
+[example | description | pointer to more information | …] <!-- optional -->
+
+* Good, because [argument a]
+* Good, because [argument b]
+* Bad, because [argument c]
+* … <!-- numbers of pros and cons can vary -->
+
+### [option 2]
+
+[example | description | pointer to more information | …] <!-- optional -->
+
+* Good, because [argument a]
+* Good, because [argument b]
+* Bad, because [argument c]
+* … <!-- numbers of pros and cons can vary -->
+
+### [option 3]
+
+[example | description | pointer to more information | …] <!-- optional -->
+
+* Good, because [argument a]
+* Good, because [argument b]
+* Bad, because [argument c]
+* … <!-- numbers of pros and cons can vary -->
+
+## Links <!-- optional -->
+
+* [Link type] [Link to ADR] <!-- example: Refined by [ADR-0005](0005-example.md) -->
+* … <!-- numbers of links can vary -->