@windyroad/risk-scorer 0.17.4 → 0.18.0

package/hooks/test/risk-gate.bats

@@ -1,330 +0,0 @@
-#!/usr/bin/env bats
-# Tests for .claude/hooks/lib/risk-gate.sh
-
-setup() {
-  HOOKS_DIR="$(cd "$(dirname "$BATS_TEST_FILENAME")/.." && pwd)"
-  source "$HOOKS_DIR/lib/gate-helpers.sh"
-  source "$HOOKS_DIR/lib/risk-gate.sh"
-
-  TEST_SESSION="bats-test-$$-${BATS_TEST_NUMBER}"
-  RDIR=$(_risk_dir "$TEST_SESSION")
-  SCORE_FILE="${RDIR}/commit"
-  HASH_FILE="${RDIR}/state-hash"
-
-  export RISK_TTL=5
-  rm -f "$SCORE_FILE" "$HASH_FILE"
-}
-
-teardown() {
-  rm -rf "${TMPDIR:-/tmp}/claude-risk-${TEST_SESSION}"
-  [ -n "${POLICY_DIR:-}" ] && rm -rf "$POLICY_DIR"
-  unset RISK_APPETITE 2>/dev/null || true
-}
-
-# Write a RISK-POLICY.md whose § Risk Appetite carries $1 into a fresh temp
-# dir and cd into it, so check_risk_gate reads the appetite from cwd. Pass an
-# empty string to OMIT the file entirely (absent-policy default path).
-_use_policy() {
-  local appetite_section="$1"
-  POLICY_DIR=$(mktemp -d)
-  if [ -n "$appetite_section" ]; then
-    printf '# Risk Policy\n\n## Risk Appetite\n\n%s\n\n## Impact Levels\n| L | x |\n' \
-      "$appetite_section" > "$POLICY_DIR/RISK-POLICY.md"
-  fi
-  cd "$POLICY_DIR"
-}
-
-# Helper: call check_risk_gate directly (not via run) so RISK_GATE_REASON is visible
-assert_gate_denies() {
-  local session="$1" action="$2" expected_reason="$3"
-  RISK_GATE_REASON=""
-  if check_risk_gate "$session" "$action"; then
-    echo "Expected gate to deny but it allowed"
-    return 1
-  fi
-  if [[ "$RISK_GATE_REASON" != *"$expected_reason"* ]]; then
-    echo "Expected reason to contain '$expected_reason' but got: $RISK_GATE_REASON"
-    return 1
-  fi
-}
-
-assert_gate_allows() {
-  local session="$1" action="$2"
-  if ! check_risk_gate "$session" "$action"; then
-    echo "Expected gate to allow but it denied: $RISK_GATE_REASON"
-    return 1
-  fi
-}
-
-@test "missing score file denies" {
-  assert_gate_denies "$TEST_SESSION" "commit" "No commit risk score found"
-}
-
-@test "missing-score deny instructs synchronous scorer dispatch (P402)" {
-  # A background-launched scorer's PostToolUse:Agent mark hook does not fire, so
-  # no marker persists and the gate re-blocks. The deny must tell the agent to
-  # dispatch synchronously. Mirrors external-comms-gate-canonical.bats.
-  assert_gate_denies "$TEST_SESSION" "commit" "run_in_background: false"
-}
-
-@test "score file with PENDING denies (non-numeric)" {
-  printf 'PENDING' > "$SCORE_FILE"
-  assert_gate_denies "$TEST_SESSION" "commit" "invalid value"
-}
-
-@test "score 4 allows (below threshold)" {
-  printf '4' > "$SCORE_FILE"
-  assert_gate_allows "$TEST_SESSION" "commit"
-}
-
-@test "score 5 allows (at threshold per ADR-086)" {
-  printf '5' > "$SCORE_FILE"
-  assert_gate_allows "$TEST_SESSION" "commit"
-}
-
-@test "score 8 denies (above threshold)" {
-  printf '8' > "$SCORE_FILE"
-  assert_gate_denies "$TEST_SESSION" "commit" "8/25"
-}
-
-@test "score 1 allows (very low)" {
-  printf '1' > "$SCORE_FILE"
-  assert_gate_allows "$TEST_SESSION" "commit"
-}
-
-@test "expired score file denies" {
-  printf '3' > "$SCORE_FILE"
-  # Backdate mtime by 10 seconds (TTL is 5)
-  touch -t "$(date -v-10S +%Y%m%d%H%M.%S 2>/dev/null || date -d '10 seconds ago' +%Y%m%d%H%M.%S 2>/dev/null)" "$SCORE_FILE"
-  assert_gate_denies "$TEST_SESSION" "commit" "expired"
-}
-
-@test "fresh score file allows" {
-  printf '3' > "$SCORE_FILE"
-  touch "$SCORE_FILE"
-  assert_gate_allows "$TEST_SESSION" "commit"
-}
-
-@test "drift detection: hash mismatch denies" {
-  printf '3' > "$SCORE_FILE"
-  touch "$SCORE_FILE"
-  echo "oldhash123" > "$HASH_FILE"
-  assert_gate_denies "$TEST_SESSION" "commit" "drift"
-}
-
-@test "no hash file skips drift check (backwards compat)" {
-  printf '3' > "$SCORE_FILE"
-  touch "$SCORE_FILE"
-  rm -f "$HASH_FILE"
-  assert_gate_allows "$TEST_SESSION" "commit"
-}
-
-@test "risk_gate_deny outputs valid JSON" {
-  run risk_gate_deny "Test reason"
-  [ "$status" -eq 0 ]
-  echo "$output" | python3 -c "import sys, json; json.load(sys.stdin)" 2>/dev/null
-  [[ "$output" == *"permissionDecision"* ]]
-  [[ "$output" == *"deny"* ]]
-  [[ "$output" == *"Test reason"* ]]
-}
-
-# ---------------------------------------------------------------------------
-# Three-band TTL policy (P090)
-# Band A: age < TTL/2 → pass silently, no slide
-# Band B: TTL/2 <= age < TTL → consult state-hash; if invariant, pass + slide
-#         the marker forward (touch score file) bounded by 2*TTL hard cap
-#         from the scorer-run birth time (<action>-born); if drifted, halt
-# Band C: age >= TTL → halt with existing "expired" message
-# ---------------------------------------------------------------------------
-
-# Helper: backdate file mtime by N seconds (portable between macOS and Linux)
-_backdate() {
-  local file="$1" seconds="$2"
-  local stamp
-  stamp=$(date -v-${seconds}S +%Y%m%d%H%M.%S 2>/dev/null \
-       || date -d "${seconds} seconds ago" +%Y%m%d%H%M.%S 2>/dev/null)
-  touch -t "$stamp" "$file"
-}
-
-# Helper: write a matching state-hash for the current working tree
-_write_matching_hash() {
-  local target="$1"
-  local hash
-  hash=$("$HOOKS_DIR/lib/pipeline-state.sh" --hash-inputs 2>/dev/null | _hashcmd | cut -d' ' -f1)
-  echo "$hash" > "$target"
-}
-
-@test "Band A (age < TTL/2): passes, does NOT slide the marker" {
-  printf '3' > "$SCORE_FILE"
-  touch "$SCORE_FILE"
-  rm -f "$HASH_FILE"
-  BEFORE_MTIME=$(_mtime "$SCORE_FILE")
-  sleep 1
-  assert_gate_allows "$TEST_SESSION" "commit"
-  AFTER_MTIME=$(_mtime "$SCORE_FILE")
-  [ "$BEFORE_MTIME" = "$AFTER_MTIME" ]
-}
-
-@test "Band B (TTL/2 <= age < TTL) with hash invariant: passes AND slides marker forward" {
-  printf '3' > "$SCORE_FILE"
-  _backdate "$SCORE_FILE" 3
-  _write_matching_hash "$HASH_FILE"
-  touch "${SCORE_FILE}-born"
-  BEFORE_MTIME=$(_mtime "$SCORE_FILE")
-  assert_gate_allows "$TEST_SESSION" "commit"
-  AFTER_MTIME=$(_mtime "$SCORE_FILE")
-  [ "$AFTER_MTIME" -gt "$BEFORE_MTIME" ]
-}
-
-@test "Band B with no hash file: passes but does NOT slide (no invariance proof)" {
-  printf '3' > "$SCORE_FILE"
-  _backdate "$SCORE_FILE" 2
-  rm -f "$HASH_FILE"
-  BEFORE_MTIME=$(_mtime "$SCORE_FILE")
-  sleep 1
-  assert_gate_allows "$TEST_SESSION" "commit"
-  AFTER_MTIME=$(_mtime "$SCORE_FILE")
-  [ "$BEFORE_MTIME" = "$AFTER_MTIME" ]
-}
-
-@test "Band B with hash mismatch: denies with drift (no slide)" {
-  printf '3' > "$SCORE_FILE"
-  _backdate "$SCORE_FILE" 3
-  echo "staleold" > "$HASH_FILE"
-  touch "${SCORE_FILE}-born"
-  BEFORE_MTIME=$(_mtime "$SCORE_FILE")
-  assert_gate_denies "$TEST_SESSION" "commit" "drift"
-  AFTER_MTIME=$(_mtime "$SCORE_FILE")
-  [ "$BEFORE_MTIME" = "$AFTER_MTIME" ]
-}
-
-@test "Band B with hard-cap exceeded (born-age >= 2*TTL): denies even if hash invariant" {
-  printf '3' > "$SCORE_FILE"
-  _backdate "$SCORE_FILE" 3
-  _write_matching_hash "$HASH_FILE"
-  touch "${SCORE_FILE}-born"
-  _backdate "${SCORE_FILE}-born" 12
-  assert_gate_denies "$TEST_SESSION" "commit" "expired"
-}
-
-@test "Band C (age >= TTL): denies regardless of hash invariance" {
-  printf '3' > "$SCORE_FILE"
-  _backdate "$SCORE_FILE" 10
-  _write_matching_hash "$HASH_FILE"
-  touch "${SCORE_FILE}-born"
-  assert_gate_denies "$TEST_SESSION" "commit" "expired"
-}
-
-@test "Band B denial exports RISK_GATE_CATEGORY=drift on hash mismatch" {
-  printf '3' > "$SCORE_FILE"
-  _backdate "$SCORE_FILE" 3
-  echo "staleold" > "$HASH_FILE"
-  RISK_GATE_CATEGORY=""
-  ! check_risk_gate "$TEST_SESSION" "commit"
-  [ "$RISK_GATE_CATEGORY" = "drift" ]
-}
-
-@test "Band C denial exports RISK_GATE_CATEGORY=expired" {
-  printf '3' > "$SCORE_FILE"
-  _backdate "$SCORE_FILE" 10
-  RISK_GATE_CATEGORY=""
-  ! check_risk_gate "$TEST_SESSION" "commit"
-  [ "$RISK_GATE_CATEGORY" = "expired" ]
-}
-
-@test "Threshold denial exports RISK_GATE_CATEGORY=threshold and RISK_GATE_SCORE" {
-  printf '7' > "$SCORE_FILE"
-  touch "$SCORE_FILE"
-  RISK_GATE_CATEGORY=""
-  RISK_GATE_SCORE=""
-  ! check_risk_gate "$TEST_SESSION" "commit"
-  [ "$RISK_GATE_CATEGORY" = "threshold" ]
-  [ "$RISK_GATE_SCORE" = "7" ]
-}
-
-# ---------------------------------------------------------------------------
-# P007 / ADR-065 — the block threshold is the project's RISK-POLICY.md risk
-# appetite (block when score > N), not a hardcoded 5. Default N=5 per ADR-086
-# (was 4 under superseded ADR-065) when the policy is absent or unparseable,
-# which tracks the new Low ceiling so the default fallback admits residual=5
-# (the Impact=5/Likelihood=1 floor for severe-but-rare risks). Precedence:
-# RISK_APPETITE env > RISK-POLICY.md parse > default 5. (Behavioural fixtures
-# per ADR-052.)
-# ---------------------------------------------------------------------------
-
-@test "appetite 9 (exceeds 9): score 7 within the 5-9 band PASSES" {
-  _use_policy 'Pipeline gates block when cumulative residual risk exceeds 9.'
-  printf '7' > "$SCORE_FILE"
-  rm -f "$HASH_FILE"
-  assert_gate_allows "$TEST_SESSION" "commit"
-}
-
-@test "appetite 9 (exceeds 9): score 10 above the band FAILS and deny renders the parsed appetite 9/25" {
-  _use_policy 'Pipeline gates block when cumulative residual risk exceeds 9.'
-  printf '10' > "$SCORE_FILE"
-  rm -f "$HASH_FILE"
-  assert_gate_denies "$TEST_SESSION" "commit" "appetite of 9/25"
-}
-
-@test "appetite via 'Threshold: 9' phrasing: score 9 PASSES, score 10 FAILS" {
-  _use_policy '**Threshold: 9 (Medium)**'
-  rm -f "$HASH_FILE"
-  printf '9' > "$SCORE_FILE"
-  assert_gate_allows "$TEST_SESSION" "commit"
-  printf '10' > "$SCORE_FILE"
-  assert_gate_denies "$TEST_SESSION" "commit" "10/25"
-}
-
-@test "appetite 4 (exceeds 4): score 4 PASSES, score 5 FAILS" {
-  _use_policy 'Pipeline gates block when cumulative residual risk exceeds 4.'
-  rm -f "$HASH_FILE"
-  printf '4' > "$SCORE_FILE"
-  assert_gate_allows "$TEST_SESSION" "commit"
-  printf '5' > "$SCORE_FILE"
-  assert_gate_denies "$TEST_SESSION" "commit" "appetite of 4/25"
-}
-
-@test "absent RISK-POLICY.md: defaults to appetite 5 per ADR-086 (5 PASSES, 6 FAILS)" {
-  _use_policy ''   # no RISK-POLICY.md in the temp dir
-  rm -f "$HASH_FILE"
-  printf '5' > "$SCORE_FILE"
-  assert_gate_allows "$TEST_SESSION" "commit"
-  printf '6' > "$SCORE_FILE"
-  assert_gate_denies "$TEST_SESSION" "commit" "appetite of 5/25"
-}
-
-@test "unparseable RISK-POLICY.md (no appetite integer): defaults to appetite 5 per ADR-086" {
-  _use_policy 'We are conservative about risk but state no number here.'
-  rm -f "$HASH_FILE"
-  printf '5' > "$SCORE_FILE"
-  assert_gate_allows "$TEST_SESSION" "commit"
-  printf '6' > "$SCORE_FILE"
-  assert_gate_denies "$TEST_SESSION" "commit" "appetite of 5/25"
-}
-
-@test "fractional score 5.5 FAILS under default appetite 5 (5.5 > 5) per ADR-086" {
-  _use_policy ''
-  rm -f "$HASH_FILE"
-  printf '5.5' > "$SCORE_FILE"
-  assert_gate_denies "$TEST_SESSION" "commit" "5.5/25"
-}
-
-@test "appetite via 'Threshold: 5' phrasing (ADR-086 new Low ceiling): score 5 PASSES, score 6 FAILS" {
-  _use_policy '**Threshold: 5 (Low)**'
-  rm -f "$HASH_FILE"
-  printf '5' > "$SCORE_FILE"
-  assert_gate_allows "$TEST_SESSION" "commit"
-  printf '6' > "$SCORE_FILE"
-  assert_gate_denies "$TEST_SESSION" "commit" "appetite of 5/25"
-}
-
-@test "RISK_APPETITE env override takes precedence over RISK-POLICY.md parse" {
-  _use_policy 'Pipeline gates block when cumulative residual risk exceeds 4.'
-  rm -f "$HASH_FILE"
-  RISK_APPETITE=9
-  printf '7' > "$SCORE_FILE"
-  assert_gate_allows "$TEST_SESSION" "commit"
-  printf '10' > "$SCORE_FILE"
-  assert_gate_denies "$TEST_SESSION" "commit" "appetite of 9/25"
-  unset RISK_APPETITE
-}

package/hooks/test/risk-policy-project-root.bats

@@ -1,20 +0,0 @@
-#!/usr/bin/env bats
-
-# P004: risk-policy-enforce-edit.sh project-root check.
-
-setup() {
-  SCRIPT_DIR="$(cd "$(dirname "$BATS_TEST_FILENAME")/.." && pwd)"
-  HOOK="$SCRIPT_DIR/risk-policy-enforce-edit.sh"
-}
-
-run_hook_with_file() {
-  local file_path="$1"
-  local json="{\"tool_input\":{\"file_path\":\"${file_path}\"},\"session_id\":\"test-$$\"}"
-  echo "$json" | bash "$HOOK"
-}
-
-@test "risk-policy project-root: RISK-POLICY.md outside project exits 0" {
-  run run_hook_with_file "/Users/other/project/RISK-POLICY.md"
-  [ "$status" -eq 0 ]
-  [[ "$output" != *"BLOCKED"* ]]
-}

package/hooks/test/risk-score-commit-gate-adr-031-bypass.bats

@@ -1,98 +0,0 @@
-#!/usr/bin/env bats
-
-# P170 / RFC-002 / ADR-031 T11: risk-score-commit-gate.sh recognises
-# the `RISK_BYPASS: adr-031-migration` self-attestation marker in the
-# git commit command's message arguments. Migration commits emitted by
-# `migrate_problems_to_per_state_layout` carry this marker in their
-# body so adopter auto-migration commits skip the full risk-score
-# overhead while preserving the audit trail (per ADR-031 § Open
-# Execution-time Questions resolution Q3 lean (b)).
-
-setup() {
-  HOOKS_DIR="$(cd "$(dirname "$BATS_TEST_FILENAME")/.." && pwd)"
-  GATE_SCRIPT="$HOOKS_DIR/risk-score-commit-gate.sh"
-  TEST_SESSION="bats-test-$$-${BATS_TEST_NUMBER}"
-
-  # Run inside a temp work tree so the gate's RISK-POLICY.md presence
-  # check has something to bind to. The migration test fixture (T10)
-  # creates a real adopter repo; here we just need a minimal stub.
-  TMP_REPO="$(mktemp -d)"
-  cd "$TMP_REPO"
-  cat > RISK-POLICY.md <<EOF
-# Risk Policy
-
-Last reviewed: $(date -u +%Y-%m-%d)
-EOF
-}
-
-teardown() {
-  rm -rf "$TMP_REPO"
-  rm -rf "${TMPDIR:-/tmp}/claude-risk-${TEST_SESSION}"
-}
-
-# Helper: invoke the gate with a Bash tool_input json + given command.
-# Returns 0 if gate allows (exits 0 with no JSON deny), non-zero on deny.
-invoke_gate() {
-  local command="$1"
-  local input
-  input=$(python3 -c "
-import json, sys
-print(json.dumps({
-    'tool_name': 'Bash',
-    'tool_input': {'command': sys.argv[1]},
-    'session_id': sys.argv[2]
-}))
-" "$command" "$TEST_SESSION")
-  echo "$input" | bash "$GATE_SCRIPT"
-}
-
-@test "T11: commit with RISK_BYPASS: adr-031-migration marker bypasses the gate" {
-  local cmd='git commit -m "docs(problems): auto-migrate" -m "RISK_BYPASS: adr-031-migration"'
-  run invoke_gate "$cmd"
-  [ "$status" -eq 0 ]
-  # Gate should emit NO deny JSON when bypassed (silent allow).
-  [[ "$output" != *"permissionDecision"* ]] || [[ "$output" != *"deny"* ]]
-}
-
-@test "T11: commit message body containing RISK_BYPASS marker (multi-paragraph -m sequence) bypasses" {
-  local cmd='git commit -m "docs(problems): auto-migrate to per-state subdirectory layout (ADR-031)" -m "See: docs/decisions/031-problem-ticket-directory-layout.accepted.md" -m "RISK_BYPASS: adr-031-migration"'
-  run invoke_gate "$cmd"
-  [ "$status" -eq 0 ]
-  [[ "$output" != *"permissionDecision"* ]] || [[ "$output" != *"deny"* ]]
-}
-
-@test "T11: normal commit without RISK_BYPASS marker still gated (no score = deny)" {
-  local cmd='git commit -m "feat: normal commit"'
-  run invoke_gate "$cmd"
-  # Either exits with deny JSON output OR exits non-zero — both indicate the
-  # gate didn't silently allow. We assert deny JSON appears in the output.
-  [[ "$output" == *"deny"* ]]
-}
-
-@test "T11: heredoc-style git commit -m with embedded RISK_BYPASS marker bypasses" {
-  # Heredoc shape used by the migration routine + manual workflows;
-  # marker must be detected regardless of how the message reaches the
-  # command string.
-  local cmd=$'git commit -m "$(cat <<\'EOF\'\ndocs(problems): auto-migrate\n\nRISK_BYPASS: adr-031-migration\nEOF\n)"'
-  run invoke_gate "$cmd"
-  [ "$status" -eq 0 ]
-  [[ "$output" != *"permissionDecision"* ]] || [[ "$output" != *"deny"* ]]
-}
-
-@test "T11: marker is case-sensitive (adr-031-MIGRATION not recognised — security guard)" {
-  local cmd='git commit -m "RISK_BYPASS: adr-031-MIGRATION"'
-  run invoke_gate "$cmd"
-  [[ "$output" == *"deny"* ]]
-}
-
-@test "T11: unrelated RISK_BYPASS token (e.g. reducing) does NOT match adr-031-migration path" {
-  # `reducing` is the existing risk-score-mark.sh marker for risk-reducing
-  # scoring runs — different mechanism (filesystem marker via PostToolUse
-  # agent return), not commit-message detection. The commit-gate should
-  # NOT treat a `RISK_BYPASS: reducing` line in the COMMIT MESSAGE as a
-  # bypass — only `adr-031-migration` (and any future commit-message
-  # markers added explicitly).
-  local cmd='git commit -m "RISK_BYPASS: reducing"'
-  run invoke_gate "$cmd"
-  [[ "$output" == *"deny"* ]]
-}

package/hooks/test/risk-score-commit-gate-cadence-staleness.bats

@@ -1,105 +0,0 @@
-#!/usr/bin/env bats
-
-# P408 / ADR-091 / RFC-043 / STORY-037: the POLICY_STALE branch of
-# risk-score-commit-gate.sh derives its staleness threshold from the
-# policy's stated review cadence line (`> Reviewed <cadence> ...`)
-# instead of a hardcoded 14 days. Vocabulary: weekly=7,
-# fortnightly/biweekly=14, monthly=30, quarterly=90, annually/yearly=365;
-# fallback 14 when the line is absent or the word unrecognised. The
-# cadence regex is case-sensitive and must bind the capital-R
-# `Reviewed <word>` line, never the lowercase `Last reviewed: <date>`
-# line (ADR-091 machine-read contract).
-
-setup() {
-  HOOKS_DIR="$(cd "$(dirname "$BATS_TEST_FILENAME")/.." && pwd)"
-  GATE_SCRIPT="$HOOKS_DIR/risk-score-commit-gate.sh"
-  TEST_SESSION="bats-test-$$-${BATS_TEST_NUMBER}"
-  TMP_REPO="$(mktemp -d)"
-  cd "$TMP_REPO"
-}
-
-teardown() {
-  rm -rf "$TMP_REPO"
-  rm -rf "${TMPDIR:-/tmp}/claude-risk-${TEST_SESSION}"
-}
-
-# Helper: date N days ago, ISO format.
-days_ago() {
-  python3 -c "from datetime import date, timedelta; print(date.today() - timedelta(days=int('$1')))"
-}
-
-# Helper: write a RISK-POLICY.md reviewed N days ago with an optional
-# cadence line supplied verbatim as the second argument.
-write_policy() {
-  local reviewed_days_ago="$1"
-  local cadence_line="${2-}"
-  {
-    echo "# Risk Policy"
-    echo ""
-    echo "> Last reviewed: $(days_ago "$reviewed_days_ago")"
-    if [ -n "$cadence_line" ]; then
-      echo ""
-      echo "$cadence_line"
-    fi
-  } > RISK-POLICY.md
-}
-
-# Helper: invoke the gate with a git-commit Bash payload.
-invoke_gate() {
-  local input
-  input=$(python3 -c "
-import json, sys
-print(json.dumps({
-    'tool_name': 'Bash',
-    'tool_input': {'command': 'git commit -m \"feat: change\"'},
-    'session_id': sys.argv[1]
-}))
-" "$TEST_SESSION")
-  echo "$input" | bash "$GATE_SCRIPT"
-}
-
-@test "P408: monthly cadence, reviewed 16 days ago — NOT stale (witnessed case; deny is the score gate, not staleness)" {
-  write_policy 16 "> Reviewed monthly and after any significant change to distribution channels."
-  run invoke_gate
-  # The gate still denies (no risk score for this session) but the reason
-  # must NOT be policy staleness.
-  [[ "$output" == *"deny"* ]]
-  [[ "$output" != *"stale"* ]]
-}
-
-@test "P408: no cadence line, reviewed 16 days ago — 14-day fallback still flags stale (behaviour preserved)" {
-  write_policy 16
-  run invoke_gate
-  [[ "$output" == *"stale"* ]]
-}
-
-@test "P408: unrecognised cadence word falls back to 14 days" {
-  write_policy 16 "> Reviewed sporadically whenever convenient."
-  run invoke_gate
-  [[ "$output" == *"stale"* ]]
-}
-
-@test "P408: regex binds the capital-R cadence line, never the lowercase 'Last reviewed:' date line" {
-  # Quarterly cadence, reviewed 20 days ago. If the regex wrongly bound
-  # the 'Last reviewed:' line the capture would fail and the 14-day
-  # fallback would flag stale at 20 days; correct capital-R binding
-  # derives 90 days and passes.
-  write_policy 20 "> Reviewed quarterly."
-  run invoke_gate
-  [[ "$output" == *"deny"* ]]
-  [[ "$output" != *"stale"* ]]
-}
-
-@test "P408: elapsed monthly cadence denies and the message names the derived threshold and cadence word" {
-  write_policy 45 "> Reviewed monthly."
-  run invoke_gate
-  [[ "$output" == *"stale"* ]]
-  [[ "$output" == *"30"* ]]
-  [[ "$output" == *"monthly"* ]]
-}
-
-@test "P408: weekly cadence tightens the threshold below the default (10 days is stale under weekly)" {
-  write_policy 10 "> Reviewed weekly."
-  run invoke_gate
-  [[ "$output" == *"stale"* ]]
-}

package/hooks/test/risk-score-mark-external-comms-prompt-parse.bats

@@ -1,94 +0,0 @@
-#!/usr/bin/env bats
-# Behavioural tests for risk-score-mark.sh external-comms branch under
-# P166 hook-side key derivation (ADR-028 amended 2026-05-16).
-#
-# Contract: the PostToolUse:Agent hook derives the marker key from
-# tool_input.prompt's `SURFACE: <name>` + `<draft>...</draft>` structure
-# instead of trusting an agent-emitted EXTERNAL_COMMS_RISK_KEY line.
-# On PASS, writes external-comms-risk-reviewed-<KEY> at the derived key.
-# Backward-compat: falls back to agent-emitted KEY when prompt has no
-# structure (one release-cycle window).
-
-setup() {
-  SCRIPT_DIR="$(cd "$(dirname "$BATS_TEST_FILENAME")/.." && pwd)"
-  HOOK="$SCRIPT_DIR/risk-score-mark.sh"
-  ORIG_DIR="$PWD"
-  TEST_DIR=$(mktemp -d)
-  cd "$TEST_DIR"
-  TMPDIR="$TEST_DIR/tmp"
-  export TMPDIR
-  mkdir -p "$TMPDIR"
-  SESSION_ID="test-rs-mark-extcomms-prompt-$$-${BATS_TEST_NUMBER}"
-  RDIR="$TMPDIR/claude-risk-${SESSION_ID}"
-}
-
-teardown() {
-  cd "$ORIG_DIR"
-  rm -rf "$TEST_DIR"
-}
-
-gate_key() {
-  local draft="$1" surface="$2"
-  printf '%s\n%s' "$draft" "$surface" | shasum -a 256 | cut -d' ' -f1
-}
-
-run_hook() {
-  local prompt="$1"
-  local agent_output="$2"
-  python3 -c "
-import json, sys
-print(json.dumps({
-  'tool_name': 'Agent',
-  'session_id': '${SESSION_ID}',
-  'tool_input': {'subagent_type': 'wr-risk-scorer:external-comms', 'prompt': sys.argv[1]},
-  'tool_response': {'content': [{'type': 'text', 'text': sys.argv[2]}]}
-}))" "$prompt" "$agent_output" | bash "$HOOK"
-}
-
-@test "external-comms PASS with structured prompt: marker lands at hook-derived key" {
-  DRAFT="we observed a leaked secret pattern in the changeset"
-  SURFACE="changeset-author"
-  PROMPT=$'SURFACE: '"$SURFACE"$'\n<draft>\n'"$DRAFT"$'\n</draft>\nReview against RISK-POLICY.md.'
-  AGENT_OUTPUT=$'no Confidential Information class matched\nEXTERNAL_COMMS_RISK_VERDICT: PASS'
-  run_hook "$PROMPT" "$AGENT_OUTPUT"
-  KEY=$(gate_key "$DRAFT" "$SURFACE")
-  [ -f "$RDIR/external-comms-risk-reviewed-${KEY}" ]
-}
-
-@test "external-comms FAIL with structured prompt: no marker" {
-  DRAFT="client Acme Corp is hitting this"
-  SURFACE="gh-issue-create"
-  PROMPT=$'SURFACE: '"$SURFACE"$'\n<draft>\n'"$DRAFT"$'\n</draft>'
-  AGENT_OUTPUT=$'EXTERNAL_COMMS_RISK_VERDICT: FAIL\nEXTERNAL_COMMS_RISK_REASON: Client names class — "Acme Corp"'
-  run_hook "$PROMPT" "$AGENT_OUTPUT"
-  KEY=$(gate_key "$DRAFT" "$SURFACE")
-  [ ! -f "$RDIR/external-comms-risk-reviewed-${KEY}" ]
-}
-
-@test "external-comms PASS with structured prompt AND agent-emitted KEY: hook-derived key wins" {
-  DRAFT="hook-derived wins"
-  SURFACE="gh-pr-comment"
-  PROMPT=$'SURFACE: '"$SURFACE"$'\n<draft>\n'"$DRAFT"$'\n</draft>'
-  BOGUS_KEY="0000000000000000000000000000000000000000000000000000000000000000"
-  AGENT_OUTPUT=$'EXTERNAL_COMMS_RISK_VERDICT: PASS\nEXTERNAL_COMMS_RISK_KEY: '"$BOGUS_KEY"
-  run_hook "$PROMPT" "$AGENT_OUTPUT"
-  DERIVED_KEY=$(gate_key "$DRAFT" "$SURFACE")
-  [ -f "$RDIR/external-comms-risk-reviewed-${DERIVED_KEY}" ]
-  [ ! -f "$RDIR/external-comms-risk-reviewed-${BOGUS_KEY}" ]
-}
-
-@test "external-comms backward-compat: PASS with no structured prompt but agent KEY" {
-  LEGACY_KEY="fedcba9876543210fedcba9876543210fedcba9876543210fedcba9876543210"
-  PROMPT="legacy unstructured prompt"
-  AGENT_OUTPUT=$'EXTERNAL_COMMS_RISK_VERDICT: PASS\nEXTERNAL_COMMS_RISK_KEY: '"$LEGACY_KEY"
-  run_hook "$PROMPT" "$AGENT_OUTPUT"
-  [ -f "$RDIR/external-comms-risk-reviewed-${LEGACY_KEY}" ]
-}
-
-@test "external-comms no structured prompt and no agent KEY: no marker" {
-  PROMPT="legacy"
-  AGENT_OUTPUT=$'EXTERNAL_COMMS_RISK_VERDICT: PASS'
-  run_hook "$PROMPT" "$AGENT_OUTPUT"
-  ext_markers=$(find "$RDIR" -maxdepth 1 -name 'external-comms-risk-reviewed-*' 2>/dev/null | wc -l | tr -d ' ')
-  [ "$ext_markers" -eq 0 ]
-}