claude-dev-env 2.16.0 → 2.18.0

package/.agents/skills/review-router/scripts/test_review_router.py

@@ -0,0 +1,23 @@
+from review_router import load_route_policy, resolve_route
+
+
+def test_load_route_policy_returns_versioned_policy() -> None:
+    assert load_route_policy()["version"] == 1
+
+
+def test_e_simplify_route_and_unsupported_e_code_review() -> None:
+    assert resolve_route("e-simplify", "T1")["status"] == "SUPPORTED"
+    assert resolve_route("e-code-review", "T1")["status"] == "UNSUPPORTED_ROUTE"
+
+
+def test_route_has_only_current_dispatch_fields() -> None:
+    route_result = resolve_route("e-simplify", "T1")
+    assert route_result["dispatch"]["pass_ids"] == ["simplify-01"]
+    assert "advisor" not in route_result
+    assert "same_executor_redirect" not in route_result
+
+
+def test_t3_has_six_ordered_slots() -> None:
+    assert resolve_route("e-simplify", "T3")["dispatch"]["pass_ids"] == [
+        "simplify-01", "simplify-02", "simplify-03", "simplify-04", "simplify-05", "simplify-06"
+    ]

package/.agents/skills/review-router/scripts/test_review_router_cli.py

@@ -0,0 +1,47 @@
+"""CLI production-path acceptance tests."""
+
+from __future__ import annotations
+
+import os
+import subprocess
+import sys
+from pathlib import Path
+import pytest
+
+from review_router_cli import arm, close, resolve
+from review_router_constants.config.constants import UNSUPPORTED_ROUTE
+
+CLI = Path(__file__).with_name("review_router_cli.py")
+
+
+def test_public_arm_and_close_reject_missing_signed_state(tmp_path: Path) -> None:
+    with pytest.raises(OSError):
+        arm(str(tmp_path), "missing", "slot")
+    with pytest.raises(OSError):
+        close(str(tmp_path), "missing")
+
+
+def test_resolve_returns_unsupported_for_unknown_route(tmp_path: Path) -> None:
+    resolve_response = resolve(str(tmp_path), "unknown-route", "", None)
+    assert resolve_response["status"] == UNSUPPORTED_ROUTE
+
+
+def test_cli_rejects_requested_dispatch(tmp_path: Path) -> None:
+    command_result = subprocess.run(
+        [sys.executable, str(CLI), "next", "--cwd", str(tmp_path)],
+        capture_output=True,
+        text=True,
+    )
+    assert command_result.returncode != 0
+    assert "invalid choice" in command_result.stderr
+
+
+@pytest.mark.parametrize("arguments", ["--tier 4", "--tier", "--tier x", "--tier 1 --tier 2"])
+def test_malformed_tier_override_writes_no_artifacts(tmp_path: Path, arguments: str) -> None:
+    repository = tmp_path / "repository"
+    repository.mkdir()
+    subprocess.run(["git", "init", "-q"], cwd=repository, check=True)
+    environment = {**os.environ, "CLAUDE_PLUGIN_DATA": str(tmp_path / "data")}
+    command_result = subprocess.run([sys.executable, str(CLI), "resolve", "--cwd", str(repository), "--arguments", arguments], capture_output=True, text=True, env=environment)
+    assert command_result.returncode != 0
+    assert not (tmp_path / "data").exists()

package/.agents/skills/review-tier/SKILL.md

@@ -0,0 +1,56 @@
+---
+name: review-tier
+description: >-
+  Classify `review_tier_constants` from change axes, hard triggers, and user overrides. Triggers: /review-tier, tier override, and review classification.
+---
+
+# Review Tier
+
+Classify one current change surface into T1, T2, or T3 using `reference/tier-policy.json` and `scripts/review_tier.py`.
+
+## Composition
+
+| Skill | When | Produces |
+|---|---|---|
+| `/review-router` | After classification | Current decision route and guarded execution |
+| `/task-build` | At workflow start | Registered classification tasks |
+
+If `/review-router` is unavailable, return the current classification.
+
+## When this applies
+
+Use for `/review-tier` and automatic review classification. Refuse execution requests with: `This is a route execution request — use /review-router.` Refuse malformed policy with: `The review-tier policy is invalid; repair the policy before classification.`
+
+## Process
+
+1. Register `reference/task-seeds.md` with the host task tool.
+2. Run `scripts/review_tier.py` and record the current decision fields.
+3. Hand the current decision to `/review-router` when route execution is requested.
+
+## Examples
+
+Input `{"files": 0, "lines": 0}` produces `T1`. Input `{"hard_triggers": ["migration"]}` produces `T3`. Input `{"files": 0}` with override `T2` produces calculated `T1` and effective `T2`.
+
+## Dependencies
+
+Requires Python 3.11+ and the standard library only. Treat a non-zero script exit as a failed classification.
+
+## File index
+
+| File | Purpose |
+|---|---|
+| `SKILL.md` | Classification workflow and boundaries |
+| `reference/tier-policy.json` | Versioned axes, thresholds, and hard triggers |
+| `reference/task-seeds.md` | Ordered classification tasks |
+| `scripts/review_tier.py` | Deterministic classifier |
+| `scripts/test_review_tier.py` | Classifier behavior tests |
+| `scripts/test_generation.py` | Inventory generation tests |
+| `scripts/conftest.py` | Test import setup |
+| `scripts/review_tier_constants/__init__.py` | Constants package |
+| `scripts/review_tier_constants/config/__init__.py` | Runtime configuration package |
+| `scripts/review_tier_constants/config/constants.py` | Portable tier constants |
+
+## Folder map
+
+- `reference/` — policy and task data.
+- `scripts/` — executable classifier and paired tests.

package/.agents/skills/review-tier/reference/task-seeds.md

@@ -0,0 +1 @@
+Register classification inputs, policy validation, tier selection, and override resolution as host tasks. Complete each with command output or the current decision.

package/.agents/skills/review-tier/reference/tier-policy.json

@@ -0,0 +1,9 @@
+{
+  "version": 1,
+  "tiers": ["T1", "T2", "T3"],
+  "thresholds": {"T1": 0, "T2": 2, "T3": 5},
+  "hard_triggers": ["security", "migration", "public-api", "release"],
+  "axes": ["files", "lines", "packages", "risk", "public_api", "dependencies"],
+  "axis_weights": {"files": 1, "lines": 1, "packages": 2, "risk": 2, "public_api": 2, "dependencies": 1},
+  "approved_downgrades": ["T1", "T2"]
+}

package/.agents/skills/review-tier/scripts/conftest.py

@@ -0,0 +1,4 @@
+from pathlib import Path
+import sys
+
+sys.path.insert(0, str(Path(__file__).resolve().parent))

package/.agents/skills/review-tier/scripts/review_tier.py

@@ -0,0 +1,332 @@
+"""Build canonical pull-request evidence and select a review tier."""
+
+from __future__ import annotations
+
+import hashlib
+import json
+import os
+import subprocess
+from pathlib import Path
+from typing import Mapping, Sequence
+
+from review_tier_constants.config.constants import ALL_DEPENDENCY_MARKERS, ALL_HARD_TRIGGER_MARKERS, ALL_PACKAGE_ROOT_NAMES, ALL_PUBLIC_API_MARKERS, ALL_SOURCE_SUFFIXES, ALL_STATUS_ARGUMENTS, ALL_STATUS_DOMAINS, ALL_TIER_ORDER, ALL_UNTRACKED_ARGUMENTS, AMBIGUOUS_BASE_REF, DEFAULT_BRANCH_FALLBACK, GIT_COMMAND_FAILED, GIT_HEAD_REF, INVALID_BASE_REF, JSON_INDENT, MALFORMED_TIER_POLICY, MAX_AXIS_VALUE, MIN_NONEMPTY_RISK, PATH_SEPARATOR, PLUGIN_DATA_ENVIRONMENT, REMOTE_HEAD_REF, ROOT_PACKAGE, ROUTING_STATE_ROOT_TRACKED, STATUS_CODE_LENGTH, STATUS_PREFIX_LENGTH, UNAPPROVED_TIER_DOWNGRADE, UNKNOWN_TIER
+
+
+def router_state_root(repo_root: str | Path, all_environment: Mapping[str, str] | None = None) -> Path:
+    """Return the canonical router state root for a repository."""
+    selected_environment = all_environment or os.environ
+    plugin_root = Path(selected_environment.get(PLUGIN_DATA_ENVIRONMENT, Path(repo_root).resolve() / ".claude-plugin-data"))
+    return (plugin_root / "review-routing" / "v1").resolve()
+
+
+def router_state_directory(repo_root: str | Path, all_environment: Mapping[str, str] | None = None) -> Path:
+    """Return the repository-specific directory under the router state root."""
+    resolved_root = Path(repo_root).resolve()
+    worktree_key = hashlib.sha256(str(resolved_root).encode()).hexdigest()
+    return router_state_root(resolved_root, all_environment=all_environment) / worktree_key
+
+
+def router_state_root_id(state_root: str | Path) -> str:
+    """Return the stable identity digest for a canonical router state root."""
+    return hashlib.sha256(str(Path(state_root).resolve()).casefold().encode()).hexdigest()
+
+
+def _is_inside_state_root(repo_root: Path, relative_path: str, state_root: Path) -> bool:
+    return (repo_root / relative_path).resolve().is_relative_to(state_root)
+
+def _git(root: Path, all_arguments: Sequence[str], is_required: bool) -> bytes:
+    command = ["git", *all_arguments]
+    completed_process = subprocess.run(command, cwd=root, check=is_required, capture_output=True)
+    if completed_process.returncode and is_required:
+        raise ValueError(GIT_COMMAND_FAILED)
+    return completed_process.stdout
+
+
+def _resolve_base(root: Path, requested_base: str | None) -> tuple[str, str]:
+    if requested_base:
+        candidate = _git(root, ["rev-parse", "--verify", requested_base], is_required=False).decode().strip()
+        if not candidate:
+            raise ValueError(INVALID_BASE_REF)
+        return requested_base, candidate
+    remote_head = _git(root, ["symbolic-ref", "--quiet", REMOTE_HEAD_REF], is_required=False).decode().strip()
+    if remote_head:
+        candidate = _git(root, ["rev-parse", "--verify", remote_head], is_required=False).decode().strip()
+        if candidate:
+            return remote_head, candidate
+    fallback = _git(root, ["rev-parse", "--verify", DEFAULT_BRANCH_FALLBACK], is_required=False).decode().strip()
+    if fallback:
+        return DEFAULT_BRANCH_FALLBACK, fallback
+    raise ValueError(AMBIGUOUS_BASE_REF)
+
+
+def _status_domains(root: Path, state_root: Path | None = None) -> dict[str, set[str]]:
+    all_status_lines = _git(root, list(ALL_STATUS_ARGUMENTS), is_required=True).decode(errors="surrogateescape").split("\0")
+    domains = {domain: set() for domain in ALL_STATUS_DOMAINS}
+    for each_status_line in all_status_lines:
+        if len(each_status_line) < STATUS_PREFIX_LENGTH:
+            continue
+        path = each_status_line[STATUS_PREFIX_LENGTH:]
+        if state_root is not None and _is_inside_state_root(root, path, state_root):
+            continue
+        if each_status_line[:STATUS_CODE_LENGTH] == "??":
+            domains["untracked"].add(path)
+        else:
+            if each_status_line[0] != " ":
+                domains["staged"].add(path)
+            if each_status_line[1] != " ":
+                domains["unstaged"].add(path)
+    return domains
+
+
+def _untracked_paths(root: Path, state_root: Path | None = None) -> list[str]:
+    untracked_listing = _git(root, ALL_UNTRACKED_ARGUMENTS, is_required=True)
+    all_paths = [each_path.decode(errors="surrogateescape") for each_path in untracked_listing.split(b"\0") if each_path]
+    return [each_path for each_path in all_paths if state_root is None or not _is_inside_state_root(root, each_path, state_root)]
+
+
+def _record_content(root: Path, merge_base: str, relative_path: str) -> tuple[bytes, str]:
+    file_path = root / relative_path
+    if file_path.is_symlink():
+        content = b""
+        identity = hashlib.sha256(os.readlink(file_path).encode()).hexdigest()
+    elif file_path.is_file():
+        content = file_path.read_bytes()
+        identity = hashlib.sha256(content).hexdigest()
+    else:
+        content = _git(root, ["show", f"{merge_base}:{relative_path}"], is_required=True)
+        identity = hashlib.sha256(content).hexdigest()
+    return content, identity
+
+
+def _annotate_record(root: Path, merge_base: str, relative_path: str, all_record: dict[str, object], all_domains: dict[str, set[str]]) -> None:
+    content, identity = _record_content(root, merge_base, relative_path)
+    lower_path = relative_path.lower()
+    all_record.update({"domains": sorted([each_domain for each_domain, all_paths in all_domains.items() if relative_path in all_paths] + (["committed"] if all_record["status"] != "?" else [])), "operation_by_domain": {}, "content_sha256": identity, "binary": b"\0" in content, "added": 0, "deleted": 0, "package": _package_key(relative_path), "public_api": int(any(marker.lower() in lower_path for marker in ALL_PUBLIC_API_MARKERS)), "dependency": int(any(marker.lower() in lower_path for marker in ALL_DEPENDENCY_MARKERS)), "call_path": int(lower_path.endswith(ALL_SOURCE_SUFFIXES))})
+    all_record["operation_by_domain"] = {each_domain: ("A" if all_record["status"] == "?" else all_record["status"]) for each_domain in all_record["domains"]}
+
+
+def _effective_records(root: Path, merge_base: str, all_domains: dict[str, set[str]], state_root: Path | None = None) -> list[dict[str, object]]:
+    status_bytes = _git(root, ["diff", "--name-status", "-z", merge_base, "--"], is_required=True)
+    status_tokens = status_bytes.split(b"\0")
+    records_by_path: dict[str, dict[str, object]] = {}
+    all_untracked_paths = _untracked_paths(root, state_root)
+    index = 0
+    while index < len(status_tokens) - 1:
+        status = status_tokens[index].decode(errors="surrogateescape")
+        index += 1
+        if not status:
+            continue
+        prior_path = None
+        if status[0] in {"R", "C"}:
+            prior_path = status_tokens[index].decode(errors="surrogateescape")
+            index += 1
+        current_path = status_tokens[index].decode(errors="surrogateescape")
+        index += 1
+        records_by_path[current_path] = {"path": current_path, "prior_path": prior_path, "status": status[0]}
+    for each_path in all_untracked_paths:
+        records_by_path.setdefault(each_path, {"path": each_path, "prior_path": None, "status": "?"})
+    if state_root is not None:
+        tracked_state_paths = [each_path for each_path in records_by_path if _is_inside_state_root(root, each_path, state_root) and records_by_path[each_path]["status"] != "?"]
+        if tracked_state_paths:
+            raise ValueError(ROUTING_STATE_ROOT_TRACKED)
+        records_by_path = {each_path: record for each_path, record in records_by_path.items() if not _is_inside_state_root(root, each_path, state_root)}
+    for each_path, each_record in records_by_path.items():
+        _annotate_record(root, merge_base, each_path, each_record, all_domains)
+    numstat = _git(root, ["diff", "--numstat", "-z", merge_base, "--"], is_required=True).split(b"\0")
+    index = 0
+    while index + 1 < len(numstat):
+        fields = numstat[index].split(b"\t")
+        index += 1
+        if len(fields) < STATUS_CODE_LENGTH:
+            continue
+        if len(fields) == STATUS_CODE_LENGTH:
+            if index >= len(numstat):
+                continue
+            path = numstat[index].decode(errors="surrogateescape")
+            index += 1
+        else:
+            path = fields[2].decode(errors="surrogateescape")
+        record = records_by_path.get(path)
+        if record is None:
+            continue
+        record["added"] = int(fields[0]) if fields[0].isdigit() else 0
+        record["deleted"] = int(fields[1]) if fields[1].isdigit() else 0
+        record["binary"] = fields[0] == b"-" or fields[1] == b"-"
+    for each_path in all_untracked_paths:
+        record = records_by_path[each_path]
+        content = (root / each_path).read_bytes() if (root / each_path).is_file() else b""
+        if b"\0" not in content:
+            record["added"] = content.count(b"\n")
+    return [records_by_path[each_path] for each_path in sorted(records_by_path)]
+
+
+def _committed_paths(root: Path, merge_base: str) -> set[str]:
+    committed_paths = _git(root, ["diff", "--name-only", "-z", f"{merge_base}..{GIT_HEAD_REF}"], is_required=True).decode(errors="surrogateescape")
+    return {path for path in committed_paths.split("\0") if path}
+
+
+def _hard_triggers(all_paths: set[str]) -> list[str]:
+    lower_paths = {path.lower() for path in all_paths}
+    return [name for name, markers in ALL_HARD_TRIGGER_MARKERS.items() if any(marker in path for path in lower_paths for marker in markers)]
+
+
+def inventory_generation(repo_root: str | Path, base_ref: str | None = None, state_root: str | Path | None = None) -> dict[str, object]:
+    """Return one hashed inventory covering committed and working-tree changes.
+
+    Args:
+        repo_root: Repository to inspect.
+        base_ref: Optional validated base reference.
+    Returns:
+        Canonical evidence and domain metadata.
+    """
+    root = Path(repo_root).resolve()
+    state_root = Path(state_root).resolve() if state_root is not None else None
+    head = _git(root, ["rev-parse", "--verify", GIT_HEAD_REF], is_required=True).decode().strip()
+    requested_base, base = _resolve_base(root, base_ref)
+    merge_base = _git(root, ["merge-base", base, head], is_required=True)
+    merge_base = merge_base.decode().strip()
+    domains = _status_domains(root, state_root=state_root)
+    committed = _committed_paths(root, merge_base)
+    all_paths = committed | set().union(*domains.values())
+    records = _effective_records(root, merge_base, domains, state_root=state_root)
+    packages = {str(record["package"]) for record in records}
+    public_api = int(any(record["public_api"] for record in records))
+    dependencies = int(any(record["dependency"] for record in records))
+    triggers = _hard_triggers(all_paths)
+    patch_bytes = _git(root, ["diff", "--binary", "--full-index", "--no-ext-diff", merge_base, "--"], is_required=True)
+    untracked_identities = [{"path": each_path, "sha256": record["content_sha256"]} for each_path, record in ((record["path"], record) for record in records) if "untracked" in record["domains"]]
+    base_source = "explicit" if base_ref else ("remote_default" if requested_base == REMOTE_HEAD_REF else "configured_fallback")
+    canonical = json.dumps({"base_ref": requested_base, "base_source": base_source, "base_sha": base, "merge_base_sha": merge_base, "head_sha": head, "domains": {each_domain: sorted(all_paths) for each_domain, all_paths in domains.items()}, "files": records, "untracked_identities": sorted(untracked_identities, key=lambda record: record["path"])}, sort_keys=True, separators=(",", ":")).encode() + patch_bytes
+    lines = sum(int(record["added"]) + int(record["deleted"]) for record in records)
+    digest = hashlib.sha256(canonical).hexdigest()
+    return {"paths": [record["path"] for record in records], "files_by_domain": records, "committed": sorted(committed), "staged": sorted(domains["staged"]), "unstaged": sorted(domains["unstaged"]), "untracked": sorted(domains["untracked"]), "base_ref": requested_base, "base_source": base_source, "base_sha": base, "merge_base_sha": merge_base, "head_sha": head, "merge_base": merge_base, "HEAD": head, "inventory_hash": digest, "diff_hash": digest, "files": len(records), "lines": lines, "packages": len(packages), "risk": MIN_NONEMPTY_RISK if records else 0, "public_api": public_api, "dependencies": dependencies, "hard_triggers": triggers}
+
+
+def _package_key(path: str) -> str:
+    parts = path.split("/")
+    if len(parts) > 1 and parts[0] in ALL_PACKAGE_ROOT_NAMES:
+        return PATH_SEPARATOR.join(parts[:2])
+    return parts[0] if len(parts) > 1 else ROOT_PACKAGE
+
+
+def build_generation(repo_root: str | Path, requested_override: str | None = None, card_path: str | Path | None = None, base_ref: str | None = None, all_policy: Mapping[str, object] | None = None, state_root: str | Path | None = None) -> dict[str, object]:
+    """Build and optionally persist one immutable tier card.
+
+    Args:
+        repo_root: Repository to inspect.
+        requested_override: Optional requested tier.
+        card_path: Optional destination for the card.
+        base_ref: Optional base reference.
+        all_policy: Optional tier policy.
+    Returns:
+        The generated decision card.
+    """
+    inventory = inventory_generation(repo_root, base_ref=base_ref, state_root=state_root)
+    card = build_decision(inventory, str(inventory["base_ref"]), str(inventory["HEAD"]), str(Path(repo_root).resolve()), str(inventory["inventory_hash"]), requested_override=requested_override, all_policy=all_policy)
+    card["inventory_hash"] = inventory["inventory_hash"]
+    if card_path is not None:
+        destination = Path(card_path)
+        destination.parent.mkdir(parents=True, exist_ok=True)
+        with destination.open("x", encoding="utf-8") as card_file:
+            json.dump(card, card_file, sort_keys=True, indent=JSON_INDENT)
+            card_file.write("\n")
+    return card
+
+
+def load_policy() -> dict[str, object]:
+    """Load and validate versioned tier policy data.
+
+    Args:
+    Returns:
+        Parsed policy mapping.
+    Raises:
+        ValueError: If the policy schema is invalid.
+    """
+    policy_path = Path(__file__).parent.parent / "reference" / "tier-policy.json"
+    policy = json.loads(policy_path.read_text(encoding="utf-8"))
+    if policy.get("version") != 1 or tuple(policy.get("tiers", ())) != ALL_TIER_ORDER or set(policy.get("thresholds", {})) != set(ALL_TIER_ORDER):
+        raise ValueError(MALFORMED_TIER_POLICY)
+    return policy
+
+
+def policy_hash(all_policy: Mapping[str, object]) -> str:
+    """Return the stable policy digest used by decision records.
+
+    Args:
+        all_policy: Parsed tier policy.
+    Returns:
+        Stable SHA-256 digest.
+    """
+    return canonical_json_hash(all_policy)
+
+
+def canonical_json_hash(all_record: Mapping[str, object]) -> str:
+    """Return the stable SHA-256 digest for a JSON-compatible mapping.
+
+    Args:
+        all_record: Mapping to encode.
+    Returns:
+        Stable SHA-256 digest.
+    """
+    encoded_record = json.dumps(all_record, sort_keys=True, separators=(",", ":")).encode()
+    return hashlib.sha256(encoded_record).hexdigest()
+
+
+def calculate_tier(all_evidence: Mapping[str, object], all_policy: Mapping[str, object] | None = None) -> str:
+    """Calculate a tier from six bounded axes and hard triggers.
+
+    Args:
+        all_evidence: Bounded change evidence.
+        all_policy: Optional tier policy.
+    Returns:
+        Automatic tier identifier.
+    """
+    selected_policy = all_policy or load_policy()
+    if set(all_evidence.get("hard_triggers", ())).intersection(selected_policy["hard_triggers"]):
+        return "T3"
+    score = sum(min(MAX_AXIS_VALUE, int(all_evidence.get(axis, 0))) * int(weight) for axis, weight in selected_policy["axis_weights"].items())
+    if score >= selected_policy["thresholds"]["T3"]:
+        return "T3"
+    if score >= selected_policy["thresholds"]["T2"]:
+        return "T2"
+    return "T1"
+
+
+def effective_tier(calculated_tier: str, requested_override: str | None, all_policy: Mapping[str, object] | None = None) -> str:
+    """Apply an approved explicit tier override.
+
+    Args:
+        calculated_tier: Automatic tier.
+        requested_override: Optional requested tier.
+        all_policy: Optional tier policy.
+    Returns:
+        Effective tier identifier.
+    Raises:
+        ValueError: If either tier is invalid or the downgrade is unapproved.
+    """
+    selected_policy = all_policy or load_policy()
+    if calculated_tier not in ALL_TIER_ORDER or requested_override not in (*ALL_TIER_ORDER, None):
+        raise ValueError(UNKNOWN_TIER)
+    if requested_override is None or requested_override == calculated_tier or requested_override > calculated_tier:
+        return requested_override or calculated_tier
+    if requested_override in selected_policy["approved_downgrades"]:
+        return requested_override
+    raise ValueError(UNAPPROVED_TIER_DOWNGRADE)
+
+
+def build_decision(all_change: Mapping[str, object], base: str, commit: str, worktree: str, diff_hash: str, requested_override: str | None = None, all_policy: Mapping[str, object] | None = None) -> dict[str, object]:
+    """Build a serializable tier card with evidence and policy identity.
+
+    Args:
+        all_change: Canonical inventory evidence.
+        base: Resolved base reference.
+        commit: Current HEAD.
+        worktree: Repository path.
+        diff_hash: Canonical evidence digest.
+        requested_override: Optional requested tier.
+    Returns:
+        Serializable decision card.
+    """
+    selected_policy = all_policy or load_policy()
+    calculated = calculate_tier(all_change, all_policy=selected_policy)
+    return {"calculated_tier": calculated, "requested_override": requested_override, "effective_tier": effective_tier(calculated, requested_override, all_policy=selected_policy), "evidence": dict(all_change), "policy_hash": policy_hash(selected_policy), "base_ref": base, "base_source": all_change.get("base_source"), "merge_base": all_change.get("merge_base"), "HEAD": commit, "worktree": str(Path(worktree).resolve()), "diff_hash": diff_hash}

package/.agents/skills/review-tier/scripts/review_tier_constants/__init__.py

@@ -0,0 +1 @@
+"""Review-tier constants package."""

package/.agents/skills/review-tier/scripts/review_tier_constants/config/constants.py

@@ -0,0 +1,29 @@
+"""Constants for review_tier_constants."""
+
+DEFAULT_BRANCH_FALLBACK = "main"
+GIT_HEAD_REF = "HEAD"
+REMOTE_HEAD_REF = "refs/remotes/origin/HEAD"
+MAX_AXIS_VALUE = 2
+MIN_NONEMPTY_RISK = 1
+GIT_COMMAND_FAILED = "GIT_COMMAND_FAILED"
+INVALID_BASE_REF = "INVALID_BASE_REF"
+AMBIGUOUS_BASE_REF = "AMBIGUOUS_BASE_REF"
+MALFORMED_TIER_POLICY = "MALFORMED_TIER_POLICY"
+UNKNOWN_TIER = "UNKNOWN_TIER"
+UNAPPROVED_TIER_DOWNGRADE = "UNAPPROVED_TIER_DOWNGRADE"
+ALL_TIER_ORDER = ("T1", "T2", "T3")
+ALL_PUBLIC_API_MARKERS = ("__init__.py", "api.py", "cli.py", "SKILL.md")
+ALL_DEPENDENCY_MARKERS = ("package.json", "requirements", "pyproject.toml", "poetry.lock")
+ALL_HARD_TRIGGER_MARKERS = {"security": ("security", "auth", "permission"), "migration": ("migration", "migrations"), "public-api": ("api.py", "__init__.py"), "release": ("release", "version")}
+ALL_STATUS_DOMAINS = ("staged", "unstaged", "untracked")
+JSON_INDENT = 2
+ALL_STATUS_ARGUMENTS = ("status", "--porcelain=v1", "-z", "--untracked-files=all")
+ROOT_PACKAGE = "."
+STATUS_PREFIX_LENGTH = 3
+STATUS_CODE_LENGTH = 2
+ALL_PACKAGE_ROOT_NAMES = ("packages", "apps", "libs", "services")
+PATH_SEPARATOR = "/"
+ALL_SOURCE_SUFFIXES = (".py", ".js", ".ts")
+ALL_UNTRACKED_ARGUMENTS = ("ls-files", "--others", "--exclude-standard", "-z")
+ROUTING_STATE_ROOT_TRACKED = "ROUTING_STATE_ROOT_TRACKED"
+PLUGIN_DATA_ENVIRONMENT = "CLAUDE_PLUGIN_DATA"