claudecode-omc 5.5.2 → 5.6.0

package/bundled/upstream/ecc/skills/skill-comply/tests/test_grader.py

@@ -0,0 +1,197 @@
+"""Tests for grader module — compliance scoring with LLM classification."""
+
+from pathlib import Path
+from unittest.mock import patch
+
+import pytest
+
+from scripts.grader import ComplianceResult, StepResult, grade
+from scripts.parser import ComplianceSpec, Detector, ObservationEvent, Step, parse_spec, parse_trace
+
+FIXTURES = Path(__file__).parent.parent / "fixtures"
+
+
+@pytest.fixture
+def tdd_spec():
+    return parse_spec(FIXTURES / "tdd_spec.yaml")
+
+
+@pytest.fixture
+def compliant_trace():
+    return parse_trace(FIXTURES / "compliant_trace.jsonl")
+
+
+@pytest.fixture
+def noncompliant_trace():
+    return parse_trace(FIXTURES / "noncompliant_trace.jsonl")
+
+
+def _mock_compliant_classification(spec, trace, model="haiku"):  # noqa: ARG001
+    """Simulate LLM correctly classifying a compliant trace."""
+    return {
+        "write_test": [0],
+        "run_test_red": [1],
+        "write_impl": [2],
+        "run_test_green": [3],
+        "refactor": [4],
+    }
+
+
+def _mock_noncompliant_classification(spec, trace, model="haiku"):
+    """Simulate LLM classifying a noncompliant trace (impl before test)."""
+    return {
+        "write_impl": [0],    # src/fib.py written first
+        "write_test": [1],    # test written second
+        "run_test_green": [2],  # only a passing test run
+    }
+
+
+def _mock_empty_classification(spec, trace, model="haiku"):
+    return {}
+
+
+class TestGradeCompliant:
+    @patch("scripts.grader.classify_events", side_effect=_mock_compliant_classification)
+    def test_returns_compliance_result(self, mock_cls, tdd_spec, compliant_trace) -> None:
+        result = grade(tdd_spec, compliant_trace)
+        assert isinstance(result, ComplianceResult)
+
+    @patch("scripts.grader.classify_events", side_effect=_mock_compliant_classification)
+    def test_full_compliance(self, mock_cls, tdd_spec, compliant_trace) -> None:
+        result = grade(tdd_spec, compliant_trace)
+        assert result.compliance_rate == 1.0
+
+    @patch("scripts.grader.classify_events", side_effect=_mock_compliant_classification)
+    def test_all_required_steps_detected(self, mock_cls, tdd_spec, compliant_trace) -> None:
+        result = grade(tdd_spec, compliant_trace)
+        required_results = [s for s in result.steps if s.step_id in
+                           ("write_test", "run_test_red", "write_impl", "run_test_green")]
+        assert all(s.detected for s in required_results)
+
+    @patch("scripts.grader.classify_events", side_effect=_mock_compliant_classification)
+    def test_optional_step_detected(self, mock_cls, tdd_spec, compliant_trace) -> None:
+        result = grade(tdd_spec, compliant_trace)
+        refactor = next(s for s in result.steps if s.step_id == "refactor")
+        assert refactor.detected is True
+
+    @patch("scripts.grader.classify_events", side_effect=_mock_compliant_classification)
+    def test_no_hook_promotion_recommended(self, mock_cls, tdd_spec, compliant_trace) -> None:
+        result = grade(tdd_spec, compliant_trace)
+        assert result.recommend_hook_promotion is False
+
+    @patch("scripts.grader.classify_events", side_effect=_mock_compliant_classification)
+    def test_step_evidence_not_empty(self, mock_cls, tdd_spec, compliant_trace) -> None:
+        result = grade(tdd_spec, compliant_trace)
+        for step in result.steps:
+            if step.detected:
+                assert len(step.evidence) > 0
+
+
+class TestGradeNoncompliant:
+    @patch("scripts.grader.classify_events", side_effect=_mock_noncompliant_classification)
+    def test_low_compliance(self, mock_cls, tdd_spec, noncompliant_trace) -> None:
+        result = grade(tdd_spec, noncompliant_trace)
+        assert result.compliance_rate < 1.0
+
+    @patch("scripts.grader.classify_events", side_effect=_mock_noncompliant_classification)
+    def test_write_test_fails_ordering(self, mock_cls, tdd_spec, noncompliant_trace) -> None:
+        """write_test has before_step=write_impl, but test is written AFTER impl."""
+        result = grade(tdd_spec, noncompliant_trace)
+        write_test = next(s for s in result.steps if s.step_id == "write_test")
+        assert write_test.detected is False
+
+    @patch("scripts.grader.classify_events", side_effect=_mock_noncompliant_classification)
+    def test_run_test_red_not_detected(self, mock_cls, tdd_spec, noncompliant_trace) -> None:
+        result = grade(tdd_spec, noncompliant_trace)
+        run_red = next(s for s in result.steps if s.step_id == "run_test_red")
+        assert run_red.detected is False
+
+    @patch("scripts.grader.classify_events", side_effect=_mock_noncompliant_classification)
+    def test_hook_promotion_recommended(self, mock_cls, tdd_spec, noncompliant_trace) -> None:
+        result = grade(tdd_spec, noncompliant_trace)
+        assert result.recommend_hook_promotion is True
+
+    @patch("scripts.grader.classify_events", side_effect=_mock_noncompliant_classification)
+    def test_failure_reasons_present(self, mock_cls, tdd_spec, noncompliant_trace) -> None:
+        result = grade(tdd_spec, noncompliant_trace)
+        failed_steps = [s for s in result.steps if not s.detected and s.step_id != "refactor"]
+        for step in failed_steps:
+            assert step.failure_reason is not None
+
+
+class TestGradeEdgeCases:
+    @patch("scripts.grader.classify_events", side_effect=_mock_empty_classification)
+    def test_empty_trace(self, mock_cls, tdd_spec) -> None:
+        result = grade(tdd_spec, [])
+        assert result.compliance_rate == 0.0
+        assert result.recommend_hook_promotion is True
+
+    @patch("scripts.grader.classify_events", side_effect=_mock_compliant_classification)
+    def test_compliance_rate_is_ratio_of_required_only(self, mock_cls, tdd_spec, compliant_trace) -> None:
+        result = grade(tdd_spec, compliant_trace)
+        assert result.compliance_rate == 1.0
+
+    @patch("scripts.grader.classify_events", side_effect=_mock_compliant_classification)
+    def test_spec_id_in_result(self, mock_cls, tdd_spec, compliant_trace) -> None:
+        result = grade(tdd_spec, compliant_trace)
+        assert result.spec_id == "tdd-workflow"
+
+    @patch("scripts.grader.classify_events")
+    def test_after_step_can_reference_later_declared_spec_step(self, mock_cls) -> None:
+        spec = ComplianceSpec(
+            id="out-of-order-after-step",
+            name="Out of order after_step",
+            source_rule="rules/common/testing.md",
+            version="1.0",
+            steps=(
+                Step(
+                    id="step_a",
+                    description="Occurs after step_b even though it is declared first",
+                    required=True,
+                    detector=Detector(
+                        description="Event A",
+                        after_step="step_b",
+                    ),
+                ),
+                Step(
+                    id="step_b",
+                    description="Reference step declared later",
+                    required=True,
+                    detector=Detector(
+                        description="Event B",
+                    ),
+                ),
+            ),
+            threshold_promote_to_hook=0.5,
+        )
+        trace = [
+            ObservationEvent(
+                timestamp="2026-03-20T10:00:01Z",
+                event="tool_complete",
+                tool="Write",
+                session="sess-order",
+                input='{"file_path":"src/b.py"}',
+                output="step b",
+            ),
+            ObservationEvent(
+                timestamp="2026-03-20T10:00:02Z",
+                event="tool_complete",
+                tool="Write",
+                session="sess-order",
+                input='{"file_path":"src/a.py"}',
+                output="step a",
+            ),
+        ]
+        mock_cls.return_value = {
+            "step_a": [1],
+            "step_b": [0],
+        }
+
+        result = grade(spec, trace)
+
+        step_a = next(step for step in result.steps if step.step_id == "step_a")
+        step_b = next(step for step in result.steps if step.step_id == "step_b")
+        assert step_a.detected is True
+        assert step_a.failure_reason is None
+        assert step_b.detected is True
+        assert result.compliance_rate == 1.0

package/bundled/upstream/ecc/skills/skill-comply/tests/test_parser.py

@@ -0,0 +1,90 @@
+"""Tests for parser module — JSONL trace and YAML spec parsing."""
+
+from pathlib import Path
+
+import pytest
+
+from scripts.parser import (
+    ComplianceSpec,
+    Detector,
+    ObservationEvent,
+    Step,
+    parse_spec,
+    parse_trace,
+)
+
+FIXTURES = Path(__file__).parent.parent / "fixtures"
+
+
+class TestParseTrace:
+    def test_parses_compliant_trace(self) -> None:
+        events = parse_trace(FIXTURES / "compliant_trace.jsonl")
+        assert len(events) == 5
+        assert all(isinstance(e, ObservationEvent) for e in events)
+
+    def test_events_sorted_by_timestamp(self) -> None:
+        events = parse_trace(FIXTURES / "compliant_trace.jsonl")
+        timestamps = [e.timestamp for e in events]
+        assert timestamps == sorted(timestamps)
+
+    def test_event_fields(self) -> None:
+        events = parse_trace(FIXTURES / "compliant_trace.jsonl")
+        first = events[0]
+        assert first.tool == "Write"
+        assert first.session == "sess-001"
+        assert "test_fib.py" in first.input
+        assert first.output == "File created"
+
+    def test_parses_noncompliant_trace(self) -> None:
+        events = parse_trace(FIXTURES / "noncompliant_trace.jsonl")
+        assert len(events) == 3
+        assert "src/fib.py" in events[0].input
+
+    def test_empty_file_returns_empty_list(self, tmp_path: Path) -> None:
+        empty = tmp_path / "empty.jsonl"
+        empty.write_text("")
+        events = parse_trace(empty)
+        assert events == []
+
+    def test_nonexistent_file_raises(self) -> None:
+        with pytest.raises(FileNotFoundError):
+            parse_trace(Path("/nonexistent/trace.jsonl"))
+
+
+class TestParseSpec:
+    def test_parses_tdd_spec(self) -> None:
+        spec = parse_spec(FIXTURES / "tdd_spec.yaml")
+        assert isinstance(spec, ComplianceSpec)
+        assert spec.id == "tdd-workflow"
+        assert len(spec.steps) == 5
+
+    def test_step_fields(self) -> None:
+        spec = parse_spec(FIXTURES / "tdd_spec.yaml")
+        first = spec.steps[0]
+        assert isinstance(first, Step)
+        assert first.id == "write_test"
+        assert first.required is True
+        assert isinstance(first.detector, Detector)
+        assert "test file" in first.detector.description
+        assert first.detector.before_step == "write_impl"
+
+    def test_optional_detector_fields(self) -> None:
+        spec = parse_spec(FIXTURES / "tdd_spec.yaml")
+        write_test = spec.steps[0]
+        assert write_test.detector.after_step is None
+
+        run_test_red = spec.steps[1]
+        assert run_test_red.detector.after_step == "write_test"
+        assert run_test_red.detector.before_step == "write_impl"
+
+    def test_scoring_threshold(self) -> None:
+        spec = parse_spec(FIXTURES / "tdd_spec.yaml")
+        assert spec.threshold_promote_to_hook == 0.6
+
+    def test_required_vs_optional_steps(self) -> None:
+        spec = parse_spec(FIXTURES / "tdd_spec.yaml")
+        required = [s for s in spec.steps if s.required]
+        optional = [s for s in spec.steps if not s.required]
+        assert len(required) == 4
+        assert len(optional) == 1
+        assert optional[0].id == "refactor"

package/bundled/upstream/ecc/skills/skill-stocktake/SKILL.md

@@ -0,0 +1,193 @@
+---
+description: "Use when auditing Claude skills and commands for quality. Supports Quick Scan (changed skills only) and Full Stocktake modes with sequential subagent batch evaluation."
+origin: ECC
+---
+
+# skill-stocktake
+
+Slash command (`/skill-stocktake`) that audits all Claude skills and commands using a quality checklist + AI holistic judgment. Supports two modes: Quick Scan for recently changed skills, and Full Stocktake for a complete review.
+
+## Scope
+
+The command targets the following paths **relative to the directory where it is invoked**:
+
+| Path | Description |
+|------|-------------|
+| `~/.claude/skills/` | Global skills (all projects) |
+| `{cwd}/.claude/skills/` | Project-level skills (if the directory exists) |
+
+**At the start of Phase 1, the command explicitly lists which paths were found and scanned.**
+
+### Targeting a specific project
+
+To include project-level skills, run from that project's root directory:
+
+```bash
+cd ~/path/to/my-project
+/skill-stocktake
+```
+
+If the project has no `.claude/skills/` directory, only global skills and commands are evaluated.
+
+## Modes
+
+| Mode | Trigger | Duration |
+|------|---------|---------|
+| Quick Scan | `results.json` exists (default) | 5–10 min |
+| Full Stocktake | `results.json` absent, or `/skill-stocktake full` | 20–30 min |
+
+**Results cache:** `~/.claude/skills/skill-stocktake/results.json`
+
+## Quick Scan Flow
+
+Re-evaluate only skills that have changed since the last run (5–10 min).
+
+1. Read `~/.claude/skills/skill-stocktake/results.json`
+2. Run: `bash ~/.claude/skills/skill-stocktake/scripts/quick-diff.sh \
+         ~/.claude/skills/skill-stocktake/results.json`
+   (Project dir is auto-detected from `$PWD/.claude/skills`; pass it explicitly only if needed)
+3. If output is `[]`: report "No changes since last run." and stop
+4. Re-evaluate only those changed files using the same Phase 2 criteria
+5. Carry forward unchanged skills from previous results
+6. Output only the diff
+7. Run: `bash ~/.claude/skills/skill-stocktake/scripts/save-results.sh \
+         ~/.claude/skills/skill-stocktake/results.json <<< "$EVAL_RESULTS"`
+
+## Full Stocktake Flow
+
+### Phase 1 — Inventory
+
+Run: `bash ~/.claude/skills/skill-stocktake/scripts/scan.sh`
+
+The script enumerates skill files, extracts frontmatter, and collects UTC mtimes.
+Project dir is auto-detected from `$PWD/.claude/skills`; pass it explicitly only if needed.
+Present the scan summary and inventory table from the script output:
+
+```
+Scanning:
+  ✓ ~/.claude/skills/         (17 files)
+  ✗ {cwd}/.claude/skills/    (not found — global skills only)
+```
+
+| Skill | 7d use | 30d use | Description |
+|-------|--------|---------|-------------|
+
+### Phase 2 — Quality Evaluation
+
+Launch an Agent tool subagent (**general-purpose agent**) with the full inventory and checklist:
+
+```text
+Agent(
+  subagent_type="general-purpose",
+  prompt="
+Evaluate the following skill inventory against the checklist.
+
+[INVENTORY]
+
+[CHECKLIST]
+
+Return JSON for each skill:
+{ \"verdict\": \"Keep\"|\"Improve\"|\"Update\"|\"Retire\"|\"Merge into [X]\", \"reason\": \"...\" }
+"
+)
+```
+
+The subagent reads each skill, applies the checklist, and returns per-skill JSON:
+
+`{ "verdict": "Keep"|"Improve"|"Update"|"Retire"|"Merge into [X]", "reason": "..." }`
+
+**Chunk guidance:** Process ~20 skills per subagent invocation to keep context manageable. Save intermediate results to `results.json` (`status: "in_progress"`) after each chunk.
+
+After all skills are evaluated: set `status: "completed"`, proceed to Phase 3.
+
+**Resume detection:** If `status: "in_progress"` is found on startup, resume from the first unevaluated skill.
+
+Each skill is evaluated against this checklist:
+
+```
+- [ ] Content overlap with other skills checked
+- [ ] Overlap with MEMORY.md / CLAUDE.md checked
+- [ ] Freshness of technical references verified (use WebSearch if tool names / CLI flags / APIs are present)
+- [ ] Usage frequency considered
+```
+
+Verdict criteria:
+
+| Verdict | Meaning |
+|---------|---------|
+| Keep | Useful and current |
+| Improve | Worth keeping, but specific improvements needed |
+| Update | Referenced technology is outdated (verify with WebSearch) |
+| Retire | Low quality, stale, or cost-asymmetric |
+| Merge into [X] | Substantial overlap with another skill; name the merge target |
+
+Evaluation is **holistic AI judgment** — not a numeric rubric. Guiding dimensions:
+- **Actionability**: code examples, commands, or steps that let you act immediately
+- **Scope fit**: name, trigger, and content are aligned; not too broad or narrow
+- **Uniqueness**: value not replaceable by MEMORY.md / CLAUDE.md / another skill
+- **Currency**: technical references work in the current environment
+
+**Reason quality requirements** — the `reason` field must be self-contained and decision-enabling:
+- Do NOT write "unchanged" alone — always restate the core evidence
+- For **Retire**: state (1) what specific defect was found, (2) what covers the same need instead
+  - Bad: `"Superseded"`
+  - Good: `"disable-model-invocation: true already set; superseded by continuous-learning-v2 which covers all the same patterns plus confidence scoring. No unique content remains."`
+- For **Merge**: name the target and describe what content to integrate
+  - Bad: `"Overlaps with X"`
+  - Good: `"42-line thin content; Step 4 of chatlog-to-article already covers the same workflow. Integrate the 'article angle' tip as a note in that skill."`
+- For **Improve**: describe the specific change needed (what section, what action, target size if relevant)
+  - Bad: `"Too long"`
+  - Good: `"276 lines; Section 'Framework Comparison' (L80–140) duplicates ai-era-architecture-principles; delete it to reach ~150 lines."`
+- For **Keep** (mtime-only change in Quick Scan): restate the original verdict rationale, do not write "unchanged"
+  - Bad: `"Unchanged"`
+  - Good: `"mtime updated but content unchanged. Unique Python reference explicitly imported by rules/python/; no overlap found."`
+
+### Phase 3 — Summary Table
+
+| Skill | 7d use | Verdict | Reason |
+|-------|--------|---------|--------|
+
+### Phase 4 — Consolidation
+
+1. **Retire / Merge**: present detailed justification per file before confirming with user:
+   - What specific problem was found (overlap, staleness, broken references, etc.)
+   - What alternative covers the same functionality (for Retire: which existing skill/rule; for Merge: the target file and what content to integrate)
+   - Impact of removal (any dependent skills, MEMORY.md references, or workflows affected)
+2. **Improve**: present specific improvement suggestions with rationale:
+   - What to change and why (e.g., "trim 430→200 lines because sections X/Y duplicate python-patterns")
+   - User decides whether to act
+3. **Update**: present updated content with sources checked
+4. Check MEMORY.md line count; propose compression if >100 lines
+
+## Results File Schema
+
+`~/.claude/skills/skill-stocktake/results.json`:
+
+**`evaluated_at`**: Must be set to the actual UTC time of evaluation completion.
+Obtain via Bash: `date -u +%Y-%m-%dT%H:%M:%SZ`. Never use a date-only approximation like `T00:00:00Z`.
+
+```json
+{
+  "evaluated_at": "2026-02-21T10:00:00Z",
+  "mode": "full",
+  "batch_progress": {
+    "total": 80,
+    "evaluated": 80,
+    "status": "completed"
+  },
+  "skills": {
+    "skill-name": {
+      "path": "~/.claude/skills/skill-name/SKILL.md",
+      "verdict": "Keep",
+      "reason": "Concrete, actionable, unique value for X workflow",
+      "mtime": "2026-01-15T08:30:00Z"
+    }
+  }
+}
+```
+
+## Notes
+
+- Evaluation is blind: the same checklist applies to all skills regardless of origin (ECC, self-authored, auto-extracted)
+- Archive / delete operations always require explicit user confirmation
+- No verdict branching by skill origin

package/bundled/upstream/ecc/skills/skill-stocktake/scripts/quick-diff.sh

@@ -0,0 +1,87 @@
+#!/usr/bin/env bash
+# quick-diff.sh — compare skill file mtimes against results.json evaluated_at
+# Usage: quick-diff.sh RESULTS_JSON [CWD_SKILLS_DIR]
+# Output: JSON array of changed/new files to stdout (empty [] if no changes)
+#
+# When CWD_SKILLS_DIR is omitted, defaults to $PWD/.claude/skills so the
+# script always picks up project-level skills without relying on the caller.
+#
+# Environment:
+#   SKILL_STOCKTAKE_GLOBAL_DIR   Override ~/.claude/skills (for testing only;
+#                                do not set in production — intended for bats tests)
+#   SKILL_STOCKTAKE_PROJECT_DIR  Override project dir detection (for testing only)
+
+set -euo pipefail
+
+RESULTS_JSON="${1:-}"
+CWD_SKILLS_DIR="${SKILL_STOCKTAKE_PROJECT_DIR:-${2:-$PWD/.claude/skills}}"
+GLOBAL_DIR="${SKILL_STOCKTAKE_GLOBAL_DIR:-$HOME/.claude/skills}"
+
+if [[ -z "$RESULTS_JSON" || ! -f "$RESULTS_JSON" ]]; then
+  echo "Error: RESULTS_JSON not found: ${RESULTS_JSON:-<empty>}" >&2
+  exit 1
+fi
+
+# Validate CWD_SKILLS_DIR looks like a .claude/skills path (defense-in-depth).
+# Only warn when the path exists — a nonexistent path poses no traversal risk.
+if [[ -n "$CWD_SKILLS_DIR" && -d "$CWD_SKILLS_DIR" && "$CWD_SKILLS_DIR" != */.claude/skills* ]]; then
+  echo "Warning: CWD_SKILLS_DIR does not look like a .claude/skills path: $CWD_SKILLS_DIR" >&2
+fi
+
+evaluated_at=$(jq -r '.evaluated_at' "$RESULTS_JSON")
+
+# Fail fast on a missing or malformed evaluated_at rather than producing
+# unpredictable results from ISO 8601 string comparison against "null".
+if [[ ! "$evaluated_at" =~ ^[0-9]{4}-[0-9]{2}-[0-9]{2}T[0-9]{2}:[0-9]{2}:[0-9]{2}Z$ ]]; then
+  echo "Error: invalid or missing evaluated_at in $RESULTS_JSON: $evaluated_at" >&2
+  exit 1
+fi
+
+# Pre-extract known paths from results.json once (O(1) lookup per file instead of O(n*m))
+known_paths=$(jq -r '.skills[].path' "$RESULTS_JSON" 2>/dev/null)
+
+tmpdir=$(mktemp -d)
+# Use a function to avoid embedding $tmpdir in a quoted string (prevents injection
+# if TMPDIR were crafted to contain shell metacharacters).
+_cleanup() { rm -rf "$tmpdir"; }
+trap _cleanup EXIT
+
+# Shared counter across process_dir calls — intentionally NOT local
+i=0
+
+process_dir() {
+  local dir="$1"
+  while IFS= read -r file; do
+    local mtime dp is_new
+    mtime=$(date -u -r "$file" +%Y-%m-%dT%H:%M:%SZ)
+    dp="${file/#$HOME/~}"
+
+    # Check if this file is known to results.json (exact whole-line match to
+    # avoid substring false-positives, e.g. "python-patterns" matching "python-patterns-v2").
+    if echo "$known_paths" | grep -qxF "$dp"; then
+      is_new="false"
+      # Known file: only emit if mtime changed (ISO 8601 string comparison is safe)
+      [[ "$mtime" > "$evaluated_at" ]] || continue
+    else
+      is_new="true"
+      # New file: always emit regardless of mtime
+    fi
+
+    jq -n \
+      --arg path "$dp" \
+      --arg mtime "$mtime" \
+      --argjson is_new "$is_new" \
+      '{path:$path,mtime:$mtime,is_new:$is_new}' \
+      > "$tmpdir/$i.json"
+    i=$((i+1))
+  done < <(find "$dir" -name "*.md" -type f 2>/dev/null | sort)
+}
+
+[[ -d "$GLOBAL_DIR" ]] && process_dir "$GLOBAL_DIR"
+[[ -n "$CWD_SKILLS_DIR" && -d "$CWD_SKILLS_DIR" ]] && process_dir "$CWD_SKILLS_DIR"
+
+if [[ $i -eq 0 ]]; then
+  echo "[]"
+else
+  jq -s '.' "$tmpdir"/*.json
+fi

package/bundled/upstream/ecc/skills/skill-stocktake/scripts/save-results.sh

@@ -0,0 +1,56 @@
+#!/usr/bin/env bash
+# save-results.sh — merge evaluated skills into results.json with correct UTC timestamp
+# Usage: save-results.sh RESULTS_JSON <<< "$EVAL_JSON"
+#
+# stdin format:
+#   { "skills": {...}, "mode"?: "full"|"quick", "batch_progress"?: {...} }
+#
+# Always sets evaluated_at to current UTC time via `date -u`.
+# Merges stdin .skills into existing results.json (new entries override old).
+# Optionally updates .mode and .batch_progress if present in stdin.
+
+set -euo pipefail
+
+RESULTS_JSON="${1:-}"
+
+if [[ -z "$RESULTS_JSON" ]]; then
+  echo "Error: RESULTS_JSON argument required" >&2
+  echo "Usage: save-results.sh RESULTS_JSON <<< \"\$EVAL_JSON\"" >&2
+  exit 1
+fi
+
+EVALUATED_AT=$(date -u +%Y-%m-%dT%H:%M:%SZ)
+
+# Read eval results from stdin and validate JSON before touching the results file
+input_json=$(cat)
+if ! echo "$input_json" | jq empty 2>/dev/null; then
+  echo "Error: stdin is not valid JSON" >&2
+  exit 1
+fi
+
+if [[ ! -f "$RESULTS_JSON" ]]; then
+  # Bootstrap: create new results.json from stdin JSON + current UTC timestamp
+  echo "$input_json" | jq --arg ea "$EVALUATED_AT" \
+    '. + { evaluated_at: $ea }' > "$RESULTS_JSON"
+  exit 0
+fi
+
+# Merge: new .skills override existing ones; old skills not in input_json are kept.
+# Optionally update .mode and .batch_progress if provided.
+#
+# Use mktemp for a collision-safe temp file (concurrent runs on the same RESULTS_JSON
+# would race on a predictable ".tmp" suffix; random suffix prevents silent overwrites).
+tmp=$(mktemp "${RESULTS_JSON}.XXXXXX")
+trap 'rm -f "$tmp"' EXIT
+
+jq -s \
+  --arg ea "$EVALUATED_AT" \
+  '.[0] as $existing | .[1] as $new |
+   $existing |
+   .evaluated_at = $ea |
+   .skills = ($existing.skills + ($new.skills // {})) |
+   if ($new | has("mode")) then .mode = $new.mode else . end |
+   if ($new | has("batch_progress")) then .batch_progress = $new.batch_progress else . end' \
+  "$RESULTS_JSON" <(echo "$input_json") > "$tmp"
+
+mv "$tmp" "$RESULTS_JSON"