agent-comparability 0.5.3__tar.gz → 0.5.4__tar.gz

{agent_comparability-0.5.3/src/agent_comparability.egg-info → agent_comparability-0.5.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agent-comparability
-Version: 0.5.3
+Version: 0.5.4
 Summary: Agent evaluation comparability semantics: spec + conformance suite + thin reference implementation
 License: Apache-2.0
 Project-URL: Homepage, https://github.com/Rethymus/agent-comparability
@@ -255,7 +255,7 @@ CDR — actual output:
     "rec:t1/m1/h1",
     "rec:t1/m1/h2"
   ],
-  "policy_version": "cmp/0.3.1",
+  "policy_version": "cmp/0.4.0",
   "limitations": [
     "摘要一致仅证明未同步修改，不构成抗恶意篡改（单写信任域）"
   ]
@@ -321,7 +321,7 @@ For **real Inspect EvalLog files** (not frozen cases) use the converter:
 
 ## 12. Specification & Versioning
 
-Three distinct lines: software tag (e.g. `v0.5.0`), specification `cmp/0.3.1`
+Three distinct lines: software tag (e.g. `v0.5.0`), specification `cmp/0.4.0`
 (inside SPEC), and `agent_comparability.__version__`. Judgment-table changes
 require a spec major bump (SPEC §5). See [CHANGELOG](CHANGELOG.md) for the full tag lineage
 (v0.3.1–v0.3.6 are deprecated candidates kept immutably for the record).
@@ -355,13 +355,13 @@ No — no network, no credentials, no Docker. It only ever sees the records you
 
 **Which versions should I depend on?**
 Package `v0.5.0+` (version lines unified and CI-enforced since then); specification
-`cmp/0.3.1` evolves independently — see [§12](#12-specification--versioning).
+`cmp/0.4.0` evolves independently — see [§12](#12-specification--versioning).
 
 ## Project status
 
 | Component | Status |
 |---|---|
-| Specification `cmp/0.3.1` | Stable — frozen conformance set anchors it |
+| Specification `cmp/0.4.0` | Stable — frozen conformance set anchors it |
 | Claim types | `score_comparison`, `causal`, `descriptive_registration` (+ unknown → `INCONCLUSIVE`) |
 | Conformance cases | 22 frozen pairs (A + B sets), three separated metrics, all green in CI |
 | Adapters | Inspect AI ✓ · Pydantic Evals ✓ |

{agent_comparability-0.5.3 → agent_comparability-0.5.4}/README.md

@@ -233,7 +233,7 @@ CDR — actual output:
     "rec:t1/m1/h1",
     "rec:t1/m1/h2"
   ],
-  "policy_version": "cmp/0.3.1",
+  "policy_version": "cmp/0.4.0",
   "limitations": [
     "摘要一致仅证明未同步修改，不构成抗恶意篡改（单写信任域）"
   ]
@@ -299,7 +299,7 @@ For **real Inspect EvalLog files** (not frozen cases) use the converter:
 
 ## 12. Specification & Versioning
 
-Three distinct lines: software tag (e.g. `v0.5.0`), specification `cmp/0.3.1`
+Three distinct lines: software tag (e.g. `v0.5.0`), specification `cmp/0.4.0`
 (inside SPEC), and `agent_comparability.__version__`. Judgment-table changes
 require a spec major bump (SPEC §5). See [CHANGELOG](CHANGELOG.md) for the full tag lineage
 (v0.3.1–v0.3.6 are deprecated candidates kept immutably for the record).
@@ -333,13 +333,13 @@ No — no network, no credentials, no Docker. It only ever sees the records you
 
 **Which versions should I depend on?**
 Package `v0.5.0+` (version lines unified and CI-enforced since then); specification
-`cmp/0.3.1` evolves independently — see [§12](#12-specification--versioning).
+`cmp/0.4.0` evolves independently — see [§12](#12-specification--versioning).
 
 ## Project status
 
 | Component | Status |
 |---|---|
-| Specification `cmp/0.3.1` | Stable — frozen conformance set anchors it |
+| Specification `cmp/0.4.0` | Stable — frozen conformance set anchors it |
 | Claim types | `score_comparison`, `causal`, `descriptive_registration` (+ unknown → `INCONCLUSIVE`) |
 | Conformance cases | 22 frozen pairs (A + B sets), three separated metrics, all green in CI |
 | Adapters | Inspect AI ✓ · Pydantic Evals ✓ |

{agent_comparability-0.5.3 → agent_comparability-0.5.4}/pyproject.toml

@@ -1,29 +1,29 @@
-[build-system]
-requires = ["setuptools>=68"]
-build-backend = "setuptools.build_meta"
-
-[project]
-name = "agent-comparability"
-version = "0.5.3"
-description = "Agent evaluation comparability semantics: spec + conformance suite + thin reference implementation"
-readme = "README.md"
-requires-python = ">=3.10"
-license = {text = "Apache-2.0"}
-dependencies = []
-
-[project.urls]
-Homepage = "https://github.com/Rethymus/agent-comparability"
-Repository = "https://github.com/Rethymus/agent-comparability"
-Issues = "https://github.com/Rethymus/agent-comparability/issues"
-Changelog = "https://github.com/Rethymus/agent-comparability/blob/main/CHANGELOG.md"
-
-[project.optional-dependencies]
-suite = ["pyyaml", "jsonschema"]
-inspect = ["inspect-ai"]
-pydantic = ["pydantic-evals"]
-
-[tool.setuptools.packages.find]
-where = ["src"]
-
-[tool.setuptools.package-data]
-agent_comparability = ["data/*"]
+[build-system]
+requires = ["setuptools>=68"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "agent-comparability"
+version = "0.5.4"
+description = "Agent evaluation comparability semantics: spec + conformance suite + thin reference implementation"
+readme = "README.md"
+requires-python = ">=3.10"
+license = {text = "Apache-2.0"}
+dependencies = []
+
+[project.urls]
+Homepage = "https://github.com/Rethymus/agent-comparability"
+Repository = "https://github.com/Rethymus/agent-comparability"
+Issues = "https://github.com/Rethymus/agent-comparability/issues"
+Changelog = "https://github.com/Rethymus/agent-comparability/blob/main/CHANGELOG.md"
+
+[project.optional-dependencies]
+suite = ["pyyaml", "jsonschema"]
+inspect = ["inspect-ai"]
+pydantic = ["pydantic-evals"]
+
+[tool.setuptools.packages.find]
+where = ["src"]
+
+[tool.setuptools.package-data]
+agent_comparability = ["data/*"]

agent_comparability-0.5.4/src/agent_comparability/__init__.py

@@ -0,0 +1,3 @@
+"""Agent Comparability — cmp/0.4.0 参考实现。"""
+from .core import CDR, adjudicate, POLICY_VERSION
+__version__ = "0.5.4"

{agent_comparability-0.5.3 → agent_comparability-0.5.4}/src/agent_comparability/core.py

@@ -1,16 +1,20 @@
-"""比较判定核心库 cmp/core（cmp/0.3.1 规范参考实现，第十九轮升级）。
+"""比较判定核心库 cmp/core（cmp/0.4.0 规范参考实现）。
 
+相对 0.3.1 的判定表分支修正（#27 裁决方向 B，SPEC §5 允许 minor 新增分支）：
+  ⑤score_comparison 的 INADMISSIBLE 触发移除 same_model——单变量对称性
+    （§4 行 2：仅模型不同 → ADMISSIBLE 并排）；模型-only 变化由
+    score_scale_consistent / zero_semantics_uniform 独立把关（D2 冻结对证明）。
 相对 0.2 的规范符合性修复（依据 SPEC 0.3.1；B 集冻结预期先于本修改存在）：
   ①同任务性检查入 score 判定优先级 2（B3 暴露 0.2 缺失）
   ②评分尺度一致性检查（含 0.3.1 隐式默认规则）（B4/B5 暴露 0.2 缺失）
   ③claim 路由补 registration/descriptive 关键词（B2-B5 暴露误判）
   ④policy_version 升 cmp/0.3.1（消费方按 SPEC §5 重判旧 CDR）
-判定优先级（SPEC §3.3）：INCONCLUSIVE(缺trace) > INADMISSIBLE(任务/模型/混淆/尺度) > PARTIAL(零语义) > ADMISSIBLE。
+判定优先级（SPEC §3.3）：INCONCLUSIVE(缺trace) > INADMISSIBLE(任务/混淆/尺度) > PARTIAL(零语义) > ADMISSIBLE。
 """
 
 from dataclasses import dataclass, field, asdict
 
-POLICY_VERSION = "cmp/0.3.1"
+POLICY_VERSION = "cmp/0.4.0"
 TAMPER_LIMITATION = "摘要一致仅证明未同步修改，不构成抗恶意篡改（单写信任域）"
 DEFAULT_SCALE = {"range": "0..1", "unit": "fraction", "verifier": "source"}
 KNOWN_FIELDS = {"task_id", "harness", "model", "reward", "reward_status",
@@ -92,9 +96,10 @@ def adjudicate(case: dict, requested_claim: str) -> CDR:
             cdr.admissibility = "INCONCLUSIVE"
             cdr.supported_claims.append("score_transcription(source_verifier_attested)")
             cdr.unsupported_claims.append("success_rate(含证据缺失记录)")
-        elif len(tasks) > 1 or len(models) > 1 or confounded or "score_scale_consistent" in cdr.violated_conditions:
+        elif len(tasks) > 1 or confounded or "score_scale_consistent" in cdr.violated_conditions:
+            # 0.4：same_model 移出硬触发（#27 单变量对称性）；尺度/零语义条件独立把关
             cdr.admissibility = "INADMISSIBLE"
-            cdr.unsupported_claims.append("score_comparison(任务/模型/尺度条件冲突)")
+            cdr.unsupported_claims.append("score_comparison(任务/混淆/尺度条件冲突)")
         elif any(v.startswith("zero_semantics_uniform") for v in cdr.violated_conditions):
             cdr.admissibility = "PARTIAL"
             cdr.supported_claims.append("per_status_stratified_rate(verified_zero/unresolved_zero 分列)")

{agent_comparability-0.5.3 → agent_comparability-0.5.4}/src/agent_comparability/data/SPEC.md

@@ -1,4 +1,4 @@
-# Agent Evaluation Comparability Specification — cmp/0.3.1 (Release Candidate)
+# Agent Evaluation Comparability Specification — cmp/0.4.0 (Release Candidate)
 
 规范状态：候选规范（非行业标准、无第三方认证）。本文是判定语义的单一事实来源；参考实现（`scripts/cmp/core.py`）与一切第三方实现均以本文为准，冲突时以本文裁决。可脱离任何源码阅读。
 
@@ -32,7 +32,7 @@
 | 类型 | 语义 |
 |---|---|
 | `causal` | 断言某变量（模型/Harness）**导致**了分数差异 |
-| `score_comparison` | 请求对连续分数做（并排或汇总）比较 |
+| `score_comparison` | 请求对连续分数做（并排或汇总）比较。历史冻结案例集使用早期写法 `score_compare`；参考实现通过下划线归一化与关键词匹配接受两者（实现细节，非规范语义），新代码应一律使用 `score_comparison`。 |
 | `descriptive_registration` | 仅登记条件与分数，不作比较结论 |
 | 其他 | **未支持类型** |
 
@@ -67,7 +67,10 @@
 
 **`score_comparison`**（按优先级，先命中先裁决）：
 1. 违反 `trace_available` → `INCONCLUSIVE`（来源评分转录保留于 `unknown_conditions`，禁止删除）
-2. 违反 `single_variable_isolation` **或** `same_task_id` 或 `same_model` → `INADMISSIBLE`
+2. 违反 `single_variable_isolation` **或** `same_task_id` → `INADMISSIBLE`
+   （**cmp/0.4 修正（#27）**：`same_model` 不再单独触发 INADMISSIBLE——单变量对称性，与
+   §4 行 1/行 2 一致；模型-only 变化由 `score_scale_consistent`/`zero_semantics_uniform`
+   独立把关。0.3.x 行为见版本记录。）
 3. 违反 `score_scale_consistent` → `INADMISSIBLE`
 4. 违反 `zero_semantics_uniform` → `PARTIAL`（支持按 verified/unresolved 分列的率；禁止合并 fail 率）
 5. 其余 → `ADMISSIBLE`（同尺度连续分数并排）
@@ -95,7 +98,7 @@
 
 - **未知字段**：Record 中规范未定义的字段**必须忽略**（不得报错、不得猜测），并在 `unknown_conditions` 记 `ignored_fields:[...]`。
 - **未支持 Claim**：见 §2 —— INCONCLUSIVE + `unrecognized_claim_type`，禁止近似匹配。
-- **Policy 版本**：CDR 必须携带 `policy_version`。消费方遇到 `policy_version` 不认识时**必须丢弃该 CDR 并重新判定**，不得部分信任。本规范为 `cmp/0.3.1`（0.3 的澄清版：仅补隐式默认尺度规则，判定表未变）。
+- **Policy 版本**：CDR 必须携带 `policy_version`。消费方遇到 `policy_version` 不认识时**必须丢弃该 CDR 并重新判定**，不得部分信任。本规范为 `cmp/0.4.0`（0.3→0.4：判定表分支修正——score_comparison 的 INADMISSIBLE 触发移除 `same_model`，实现单变量对称性，裁决记录见 #27；D 冻结对（D1/D2）先于实现冻结）。0.3.1 为 0.3 的澄清版（仅补隐式默认尺度规则）。
 - **规范不兼容变更**：升 minor（0.3→0.4）可新增判定分支；改判定表必须升 major 并迁移案例。
 
 ---

agent_comparability-0.5.4/src/agent_comparability/data/cases-0.4.json

@@ -0,0 +1,16 @@
+{
+  "policy_version": "cmp/0.4.0",
+  "frozen_note": "D 集（#27 裁决方向 B，先于实现冻结）：单变量对称性——模型-only 变化遵循 SPEC §4 行 2。A/B 集与既有预期零改动；D2 证明尺度条件独立把关未被放松。",
+  "cases": [
+    {"id": "D1", "category": "model-only-variation",
+     "records": [
+       {"task_id": "t-model-only", "harness": "h1", "model": "model-a", "reward": 0.8, "reward_status": "verified"},
+       {"task_id": "t-model-only", "harness": "h1", "model": "model-b", "reward": 0.6, "reward_status": "verified"}],
+     "claims": {"score_comparison": "ADMISSIBLE", "causal": "INADMISSIBLE"}},
+    {"id": "D2", "category": "model-plus-scale-violation",
+     "records": [
+       {"task_id": "t-model-scale", "harness": "h1", "model": "model-a", "reward": 0.8, "reward_status": "verified"},
+       {"task_id": "t-model-scale", "harness": "h1", "model": "model-b", "reward": 62.0, "reward_status": "verified"}],
+     "claims": {"score_comparison": "INADMISSIBLE", "descriptive_registration": "ADMISSIBLE"}}
+  ]
+}

{agent_comparability-0.5.3 → agent_comparability-0.5.4}/src/agent_comparability/data/cdr.schema.json

@@ -74,7 +74,7 @@
     },
     "policy_version": {
       "type": "string",
-      "pattern": "^cmp/0\\.3(\\.1)?$"
+      "pattern": "^cmp/0\\.[34](\\.\\d+)?$"
     },
     "limitations": {
       "type": "array",

{agent_comparability-0.5.3 → agent_comparability-0.5.4/src/agent_comparability.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agent-comparability
-Version: 0.5.3
+Version: 0.5.4
 Summary: Agent evaluation comparability semantics: spec + conformance suite + thin reference implementation
 License: Apache-2.0
 Project-URL: Homepage, https://github.com/Rethymus/agent-comparability
@@ -255,7 +255,7 @@ CDR — actual output:
     "rec:t1/m1/h1",
     "rec:t1/m1/h2"
   ],
-  "policy_version": "cmp/0.3.1",
+  "policy_version": "cmp/0.4.0",
   "limitations": [
     "摘要一致仅证明未同步修改，不构成抗恶意篡改（单写信任域）"
   ]
@@ -321,7 +321,7 @@ For **real Inspect EvalLog files** (not frozen cases) use the converter:
 
 ## 12. Specification & Versioning
 
-Three distinct lines: software tag (e.g. `v0.5.0`), specification `cmp/0.3.1`
+Three distinct lines: software tag (e.g. `v0.5.0`), specification `cmp/0.4.0`
 (inside SPEC), and `agent_comparability.__version__`. Judgment-table changes
 require a spec major bump (SPEC §5). See [CHANGELOG](CHANGELOG.md) for the full tag lineage
 (v0.3.1–v0.3.6 are deprecated candidates kept immutably for the record).
@@ -355,13 +355,13 @@ No — no network, no credentials, no Docker. It only ever sees the records you
 
 **Which versions should I depend on?**
 Package `v0.5.0+` (version lines unified and CI-enforced since then); specification
-`cmp/0.3.1` evolves independently — see [§12](#12-specification--versioning).
+`cmp/0.4.0` evolves independently — see [§12](#12-specification--versioning).
 
 ## Project status
 
 | Component | Status |
 |---|---|
-| Specification `cmp/0.3.1` | Stable — frozen conformance set anchors it |
+| Specification `cmp/0.4.0` | Stable — frozen conformance set anchors it |
 | Claim types | `score_comparison`, `causal`, `descriptive_registration` (+ unknown → `INCONCLUSIVE`) |
 | Conformance cases | 22 frozen pairs (A + B sets), three separated metrics, all green in CI |
 | Adapters | Inspect AI ✓ · Pydantic Evals ✓ |

{agent_comparability-0.5.3 → agent_comparability-0.5.4}/src/agent_comparability.egg-info/SOURCES.txt

@@ -15,6 +15,7 @@ src/agent_comparability.egg-info/requires.txt
 src/agent_comparability.egg-info/top_level.txt
 src/agent_comparability/data/SPEC.md
 src/agent_comparability/data/boundary-cases.json
+src/agent_comparability/data/cases-0.4.json
 src/agent_comparability/data/cdr.schema.json
 src/agent_comparability/data/frozen-cases-a.json
 tests/test_cli.py

{agent_comparability-0.5.3 → agent_comparability-0.5.4}/tests/test_cli.py

@@ -1,152 +1,152 @@
-"""CLI（python -m agent_comparability）正常/错误路径回归（#16）。
-
-退出码契约：
-  0 = 产出 CDR（四种 admissibility 均为裁决成功，不是程序失败）
-  2 = 输入错误（JSON 语法/契约字段缺失或类型错/文件不可读），诊断必须含出错路径
-无效输入绝不允许被转换为貌似成功的裁决。
-"""
-import json
-import subprocess
-import sys
-from pathlib import Path
-
-import pytest
-
-ROOT = Path(__file__).resolve().parents[1]
-
-VALID = {
-    "id": "cli-demo",
-    "records": [
-        {"task_id": "t1", "harness": "h1", "model": "m1", "reward": 1.0, "reward_status": "verified"},
-        {"task_id": "t1", "harness": "h2", "model": "m1", "reward": 0.5, "reward_status": "verified"},
-    ],
-}
-
-
-def run_cli(*args):
-    return subprocess.run([sys.executable, "-m", "agent_comparability", *args],
-                          capture_output=True, text=True, cwd=ROOT)
-
-
-def _tmp_case(tmp_path, payload, raw=None, name="case.json"):
-    p = tmp_path / name
-    p.write_text(raw if raw is not None else json.dumps(payload), encoding="utf-8")
-    return str(p)
-
-
-def test_valid_input_exit0_admissible(tmp_path):
-    r = run_cli(_tmp_case(tmp_path, VALID), "--claim", "score_comparison")
-    assert r.returncode == 0, r.stderr
-    cdr = json.loads(r.stdout)
-    assert cdr["admissibility"] == "ADMISSIBLE"
-    assert cdr["policy_version"] == "cmp/0.3.1"
-    assert any("抗恶意篡改" in l for l in cdr["limitations"])
-    for k in ("case_id", "requested_claim", "satisfied_conditions", "violated_conditions",
-              "unknown_conditions", "supported_claims", "unsupported_claims",
-              "evidence_refs", "comparison_scope"):
-        assert k in cdr
-
-
-@pytest.mark.parametrize("claim,expected", [
-    ("causal", "INADMISSIBLE"),
-    ("score_comparison", "ADMISSIBLE"),
-])
-def test_all_verdicts_are_exit0_not_program_failure(tmp_path, claim, expected):
-    """INADMISSIBLE/PARTIAL/INCONCLUSIVE 与 ADMISSIBLE 同为 exit 0 —— 裁决 ≠ 程序失败。"""
-    r = run_cli(_tmp_case(tmp_path, VALID), "--claim", claim)
-    assert r.returncode == 0, r.stderr
-    assert json.loads(r.stdout)["admissibility"] == expected
-
-
-def test_inconclusive_and_partial_exit0(tmp_path):
-    miss = {"id": "m", "records": [dict(VALID["records"][0], trajectory_available=False)]}
-    r = run_cli(_tmp_case(tmp_path, miss, name="miss.json"), "--claim", "score_comparison")
-    assert r.returncode == 0 and json.loads(r.stdout)["admissibility"] == "INCONCLUSIVE"
-    mixed = {"id": "x", "records": [
-        {"task_id": "t", "harness": "a", "model": "m", "reward": 0.0, "reward_status": "verified"},
-        {"task_id": "t", "harness": "b", "model": "m", "reward": 0.0, "reward_status": "unresolved_zero"}]}
-    r2 = run_cli(_tmp_case(tmp_path, mixed, name="mixed.json"), "--claim", "score_comparison")
-    assert r2.returncode == 0 and json.loads(r2.stdout)["admissibility"] == "PARTIAL"
-
-
-def test_unknown_claim_type_exit0_inconclusive(tmp_path):
-    r = run_cli(_tmp_case(tmp_path, VALID), "--claim", "leaderboard_ranking")
-    assert r.returncode == 0
-    cdr = json.loads(r.stdout)
-    assert cdr["admissibility"] == "INCONCLUSIVE"
-    assert any("unrecognized_claim_type" in u for u in cdr["unknown_conditions"])
-
-
-def test_unknown_fields_not_error_but_recorded(tmp_path):
-    """SPEC §5：未知字段必须忽略（不报错）并记入 unknown_conditions —— CLI 不得比规范更严。"""
-    payload = json.loads(json.dumps(VALID))
-    payload["records"][0]["custom_extra"] = "keep-me-ignored"
-    r = run_cli(_tmp_case(tmp_path, payload), "--claim", "score_comparison")
-    assert r.returncode == 0, r.stderr
-    cdr = json.loads(r.stdout)
-    assert cdr["admissibility"] == "ADMISSIBLE"
-    assert any("ignored_fields" in u and "custom_extra" in u for u in cdr["unknown_conditions"])
-
-
-def test_output_file_flag(tmp_path):
-    out = tmp_path / "cdr.json"
-    r = run_cli(_tmp_case(tmp_path, VALID), "--claim", "score_comparison", "-o", str(out))
-    assert r.returncode == 0, r.stderr
-    assert r.stdout.strip() == ""
-    cdr = json.loads(out.read_text(encoding="utf-8"))
-    assert cdr["admissibility"] == "ADMISSIBLE"
-
-
-# ---------- 错误路径：全部 exit 2 + 诊断含路径 ----------
-
-def test_missing_input_file_exit2(tmp_path):
-    r = run_cli(str(tmp_path / "nope.json"), "--claim", "score_comparison")
-    assert r.returncode == 2
-    assert "nope.json" in r.stderr
-
-
-def test_malformed_json_exit2(tmp_path):
-    r = run_cli(_tmp_case(tmp_path, None, raw="{not valid json"), "--claim", "score_comparison")
-    assert r.returncode == 2
-    assert "JSON" in r.stderr
-
-
-@pytest.mark.parametrize("payload,needle", [
-    ({"records": VALID["records"]}, "id"),                                   # 缺 id
-    ({"id": "x"}, "records"),                                                # 缺 records
-    ({"id": "x", "records": []}, "records"),                                 # 空 records
-    ({"id": 42, "records": VALID["records"]}, "id"),                         # id 类型错
-    ({"id": "x", "records": [{"harness": "h", "model": "m", "reward": 1.0,
-                              "reward_status": "verified"}]}, "task_id"),    # 缺必填字段
-    ({"id": "x", "records": [dict(VALID["records"][0], reward="high")]}, "reward"),        # reward 类型错
-    ({"id": "x", "records": [dict(VALID["records"][0], reward=True)]}, "reward"),          # bool 不是数值
-    ({"id": "x", "records": [dict(VALID["records"][0], reward_status=0)]}, "reward_status"),  # status 类型错
-])
-def test_contract_violations_exit2_with_path(tmp_path, payload, needle):
-    r = run_cli(_tmp_case(tmp_path, payload), "--claim", "score_comparison")
-    assert r.returncode == 2, f"期望 exit 2，实际 {r.returncode}；stdout={r.stdout!r}"
-    assert needle in r.stderr, r.stderr
-    assert "records[0]" in r.stderr or needle == "records" or needle == "id" and "records" not in r.stderr
-
-
-def test_bad_optional_field_types_exit2(tmp_path):
-    rec = dict(VALID["records"][0], trajectory_available="yes")
-    r = run_cli(_tmp_case(tmp_path, {"id": "x", "records": [rec]}), "--claim", "score_comparison")
-    assert r.returncode == 2 and "trajectory_available" in r.stderr
-    rec2 = dict(VALID["records"][0], score_scale="0..1")
-    r2 = run_cli(_tmp_case(tmp_path, {"id": "x", "records": [rec2]}, name="c2.json"),
-                 "--claim", "score_comparison")
-    assert r2.returncode == 2 and "score_scale" in r2.stderr
-    rec3 = dict(VALID["records"][0], repetition_count=0)
-    r3 = run_cli(_tmp_case(tmp_path, {"id": "x", "records": [rec3]}, name="c3.json"),
-                 "--claim", "score_comparison")
-    assert r3.returncode == 2 and "repetition_count" in r3.stderr
-
-
-def test_missing_claim_flag_exit2(tmp_path):
-    r = run_cli(_tmp_case(tmp_path, VALID))
-    assert r.returncode == 2  # argparse required 参数缺失 → 2（argparse 惯例），不得默认猜 claim
-
-
-def test_no_args_exit2():
-    assert run_cli().returncode == 2
+"""CLI（python -m agent_comparability）正常/错误路径回归（#16）。
+
+退出码契约：
+  0 = 产出 CDR（四种 admissibility 均为裁决成功，不是程序失败）
+  2 = 输入错误（JSON 语法/契约字段缺失或类型错/文件不可读），诊断必须含出错路径
+无效输入绝不允许被转换为貌似成功的裁决。
+"""
+import json
+import subprocess
+import sys
+from pathlib import Path
+
+import pytest
+
+ROOT = Path(__file__).resolve().parents[1]
+
+VALID = {
+    "id": "cli-demo",
+    "records": [
+        {"task_id": "t1", "harness": "h1", "model": "m1", "reward": 1.0, "reward_status": "verified"},
+        {"task_id": "t1", "harness": "h2", "model": "m1", "reward": 0.5, "reward_status": "verified"},
+    ],
+}
+
+
+def run_cli(*args):
+    return subprocess.run([sys.executable, "-m", "agent_comparability", *args],
+                          capture_output=True, text=True, cwd=ROOT)
+
+
+def _tmp_case(tmp_path, payload, raw=None, name="case.json"):
+    p = tmp_path / name
+    p.write_text(raw if raw is not None else json.dumps(payload), encoding="utf-8")
+    return str(p)
+
+
+def test_valid_input_exit0_admissible(tmp_path):
+    r = run_cli(_tmp_case(tmp_path, VALID), "--claim", "score_comparison")
+    assert r.returncode == 0, r.stderr
+    cdr = json.loads(r.stdout)
+    assert cdr["admissibility"] == "ADMISSIBLE"
+    assert cdr["policy_version"] == "cmp/0.4.0"
+    assert any("抗恶意篡改" in l for l in cdr["limitations"])
+    for k in ("case_id", "requested_claim", "satisfied_conditions", "violated_conditions",
+              "unknown_conditions", "supported_claims", "unsupported_claims",
+              "evidence_refs", "comparison_scope"):
+        assert k in cdr
+
+
+@pytest.mark.parametrize("claim,expected", [
+    ("causal", "INADMISSIBLE"),
+    ("score_comparison", "ADMISSIBLE"),
+])
+def test_all_verdicts_are_exit0_not_program_failure(tmp_path, claim, expected):
+    """INADMISSIBLE/PARTIAL/INCONCLUSIVE 与 ADMISSIBLE 同为 exit 0 —— 裁决 ≠ 程序失败。"""
+    r = run_cli(_tmp_case(tmp_path, VALID), "--claim", claim)
+    assert r.returncode == 0, r.stderr
+    assert json.loads(r.stdout)["admissibility"] == expected
+
+
+def test_inconclusive_and_partial_exit0(tmp_path):
+    miss = {"id": "m", "records": [dict(VALID["records"][0], trajectory_available=False)]}
+    r = run_cli(_tmp_case(tmp_path, miss, name="miss.json"), "--claim", "score_comparison")
+    assert r.returncode == 0 and json.loads(r.stdout)["admissibility"] == "INCONCLUSIVE"
+    mixed = {"id": "x", "records": [
+        {"task_id": "t", "harness": "a", "model": "m", "reward": 0.0, "reward_status": "verified"},
+        {"task_id": "t", "harness": "b", "model": "m", "reward": 0.0, "reward_status": "unresolved_zero"}]}
+    r2 = run_cli(_tmp_case(tmp_path, mixed, name="mixed.json"), "--claim", "score_comparison")
+    assert r2.returncode == 0 and json.loads(r2.stdout)["admissibility"] == "PARTIAL"
+
+
+def test_unknown_claim_type_exit0_inconclusive(tmp_path):
+    r = run_cli(_tmp_case(tmp_path, VALID), "--claim", "leaderboard_ranking")
+    assert r.returncode == 0
+    cdr = json.loads(r.stdout)
+    assert cdr["admissibility"] == "INCONCLUSIVE"
+    assert any("unrecognized_claim_type" in u for u in cdr["unknown_conditions"])
+
+
+def test_unknown_fields_not_error_but_recorded(tmp_path):
+    """SPEC §5：未知字段必须忽略（不报错）并记入 unknown_conditions —— CLI 不得比规范更严。"""
+    payload = json.loads(json.dumps(VALID))
+    payload["records"][0]["custom_extra"] = "keep-me-ignored"
+    r = run_cli(_tmp_case(tmp_path, payload), "--claim", "score_comparison")
+    assert r.returncode == 0, r.stderr
+    cdr = json.loads(r.stdout)
+    assert cdr["admissibility"] == "ADMISSIBLE"
+    assert any("ignored_fields" in u and "custom_extra" in u for u in cdr["unknown_conditions"])
+
+
+def test_output_file_flag(tmp_path):
+    out = tmp_path / "cdr.json"
+    r = run_cli(_tmp_case(tmp_path, VALID), "--claim", "score_comparison", "-o", str(out))
+    assert r.returncode == 0, r.stderr
+    assert r.stdout.strip() == ""
+    cdr = json.loads(out.read_text(encoding="utf-8"))
+    assert cdr["admissibility"] == "ADMISSIBLE"
+
+
+# ---------- 错误路径：全部 exit 2 + 诊断含路径 ----------
+
+def test_missing_input_file_exit2(tmp_path):
+    r = run_cli(str(tmp_path / "nope.json"), "--claim", "score_comparison")
+    assert r.returncode == 2
+    assert "nope.json" in r.stderr
+
+
+def test_malformed_json_exit2(tmp_path):
+    r = run_cli(_tmp_case(tmp_path, None, raw="{not valid json"), "--claim", "score_comparison")
+    assert r.returncode == 2
+    assert "JSON" in r.stderr
+
+
+@pytest.mark.parametrize("payload,needle", [
+    ({"records": VALID["records"]}, "id"),                                   # 缺 id
+    ({"id": "x"}, "records"),                                                # 缺 records
+    ({"id": "x", "records": []}, "records"),                                 # 空 records
+    ({"id": 42, "records": VALID["records"]}, "id"),                         # id 类型错
+    ({"id": "x", "records": [{"harness": "h", "model": "m", "reward": 1.0,
+                              "reward_status": "verified"}]}, "task_id"),    # 缺必填字段
+    ({"id": "x", "records": [dict(VALID["records"][0], reward="high")]}, "reward"),        # reward 类型错
+    ({"id": "x", "records": [dict(VALID["records"][0], reward=True)]}, "reward"),          # bool 不是数值
+    ({"id": "x", "records": [dict(VALID["records"][0], reward_status=0)]}, "reward_status"),  # status 类型错
+])
+def test_contract_violations_exit2_with_path(tmp_path, payload, needle):
+    r = run_cli(_tmp_case(tmp_path, payload), "--claim", "score_comparison")
+    assert r.returncode == 2, f"期望 exit 2，实际 {r.returncode}；stdout={r.stdout!r}"
+    assert needle in r.stderr, r.stderr
+    assert "records[0]" in r.stderr or needle == "records" or needle == "id" and "records" not in r.stderr
+
+
+def test_bad_optional_field_types_exit2(tmp_path):
+    rec = dict(VALID["records"][0], trajectory_available="yes")
+    r = run_cli(_tmp_case(tmp_path, {"id": "x", "records": [rec]}), "--claim", "score_comparison")
+    assert r.returncode == 2 and "trajectory_available" in r.stderr
+    rec2 = dict(VALID["records"][0], score_scale="0..1")
+    r2 = run_cli(_tmp_case(tmp_path, {"id": "x", "records": [rec2]}, name="c2.json"),
+                 "--claim", "score_comparison")
+    assert r2.returncode == 2 and "score_scale" in r2.stderr
+    rec3 = dict(VALID["records"][0], repetition_count=0)
+    r3 = run_cli(_tmp_case(tmp_path, {"id": "x", "records": [rec3]}, name="c3.json"),
+                 "--claim", "score_comparison")
+    assert r3.returncode == 2 and "repetition_count" in r3.stderr
+
+
+def test_missing_claim_flag_exit2(tmp_path):
+    r = run_cli(_tmp_case(tmp_path, VALID))
+    assert r.returncode == 2  # argparse required 参数缺失 → 2（argparse 惯例），不得默认猜 claim
+
+
+def test_no_args_exit2():
+    assert run_cli().returncode == 2