Skip to content

An agentic skills framework & software development methodology that works.

tests: annotate three kept bash tests with drill coverage notes

Jesse Vincent315ef09ebcd4
diff --git a/tests/claude-code/test-subagent-driven-development-integration.sh b/tests/claude-code/test-subagent-driven-development-integration.sh
index 95a551bca..b37862e17 100755
--- a/tests/claude-code/test-subagent-driven-development-integration.sh
+++ b/tests/claude-code/test-subagent-driven-development-integration.sh
@@ -1,6 +1,17 @@
#!/usr/bin/env bash
# Integration Test: subagent-driven-development workflow
# Actually executes a plan and verifies the new workflow behaviors
+#
+# Drill coverage: evals/scenarios/sdd-rejects-extra-features.yaml covers the
+# YAGNI enforcement subset (forbidden exports + reviewer-as-gate semantics)
+# and is stricter on that axis. This bash test additionally asserts:
+# - >=3 git commits (initial + per-task commits, exercising SDD's
+# commit-per-task workflow shape)
+# - >=2 Agent/Task subagent dispatches (drill only asserts >=1)
+# - TodoWrite usage (drill makes no assertion)
+# - test/math.test.js exists (drill relies on `npm test` succeeding)
+# - analyze-token-usage.py token-budget telemetry
+# Kept until those assertions are added to drill or explicitly retired.
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
diff --git a/tests/claude-code/test-subagent-driven-development.sh b/tests/claude-code/test-subagent-driven-development.sh
index 20d8d4c7e..d1099c118 100755
--- a/tests/claude-code/test-subagent-driven-development.sh
+++ b/tests/claude-code/test-subagent-driven-development.sh
@@ -1,6 +1,12 @@
#!/usr/bin/env bash
# Test: subagent-driven-development skill
# Verifies that the skill is loaded and follows correct workflow
+#
+# No drill coverage: this test asks the agent to *describe* SDD (string-
+# matches its verbal explanation against expected keywords like
+# "self-review", "skeptical", "worktree", "Step 1", "loop"). Drill scenarios
+# test behavior (real subagent dispatch, plan-following, review loops),
+# not description-recall. Kept by design.
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
diff --git a/tests/claude-code/test-worktree-native-preference.sh b/tests/claude-code/test-worktree-native-preference.sh
index cbfe7f293..077ea19c3 100755
--- a/tests/claude-code/test-worktree-native-preference.sh
+++ b/tests/claude-code/test-worktree-native-preference.sh
@@ -2,6 +2,11 @@
# Test: Does the agent prefer native worktree tools (EnterWorktree) over git worktree add?
# Framework: RED-GREEN-REFACTOR per testing-skills-with-subagents.md
#
+# Drill coverage: evals/scenarios/worktree-creation-under-pressure.yaml lifts
+# only the PRESSURE phase (existing .worktrees/ + urgency framing). The RED
+# and GREEN baselines below are not covered by drill — kept here so the
+# RED-GREEN-REFACTOR validation remains rerunnable end-to-end.
+#
# RED: Skill without Step 1a (no native tool preference). Agent should use git worktree add.
# GREEN: Skill with Step 1a (explicit tool naming + consent bridge). Agent should use EnterWorktree.
# PRESSURE: Same as GREEN but under time pressure with existing .worktrees/ dir.