docs+test: SKILL authoring guidance + regression tests

Adds "Writing SKILL templates" section to CLAUDE.md explaining that
templates are prompts, not scripts. Adds validation test catching
hardcoded 'main' in git commands, and resolver content test.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

CLAUDE.md

@@ -64,6 +64,23 @@ SKILL.md files are **generated** from `.tmpl` templates. To update docs:
 To add a new browse command: add it to `browse/src/commands.ts` and rebuild.
 To add a snapshot flag: add it to `SNAPSHOT_FLAGS` in `browse/src/snapshot.ts` and rebuild.
 
+## Writing SKILL templates
+
+SKILL.md.tmpl files are **prompt templates read by Claude**, not bash scripts.
+Each bash code block runs in a separate shell — variables do not persist between blocks.
+
+Rules:
+- **Use natural language for logic and state.** Don't use shell variables to pass
+  state between code blocks. Instead, tell Claude what to remember and reference
+  it in prose (e.g., "the base branch detected in Step 0").
+- **Don't hardcode branch names.** Detect `main`/`master`/etc dynamically via
+  `gh pr view` or `gh repo view`. Use `{{BASE_BRANCH_DETECT}}` for PR-targeting
+  skills. Use "the base branch" in prose, `<base>` in code block placeholders.
+- **Keep bash blocks self-contained.** Each code block should work independently.
+  If a block needs context from a previous step, restate it in the prose above.
+- **Express conditionals as English.** Instead of nested `if/elif/else` in bash,
+  write numbered decision steps: "1. If X, do Y. 2. Otherwise, do Z."
+
 ## Browser interaction
 
 When you need to interact with a browser (QA, dogfooding, cookie setup), use the