Prove planner validation commands execute without manual reconstruction

Constraint: Adding validation_commands to the planner was only useful if the emitted commands could be consumed directly, so the plan artifact needed one more layer of execution proof.
Rejected: Assume command strings are correct because they look valid | That would leave restart automation unproven.
Confidence: high
Scope-risk: narrow
Directive: Prefer executing validation_commands from the planner artifact instead of retyping equivalent checks by hand.
Tested: git diff --check; /usr/local/miniconda3/bin/python - <<'PY' ... execute validation_commands.prereq_audit and validation_commands.worker_contract_smoke from data/pgvector_eval/music20/phase1_extraction_plan_report.json ... PY
Not-tested: The remaining planner validation commands were not executed in this commit, though their sibling commands proved the artifact is directly consumable.

+{
+  "prereq_audit": {
+    "command": "cd /workspace/acr-engine && PG_DSN=\"${PG_DSN:?set PG_DSN}\" /usr/local/miniconda3/bin/python scripts/run_phase1_prereq_audit_live.py --dsn \"$PG_DSN\" --schema acr_test --output data/pgvector_eval/music20/phase1_prereq_audit_report.json",
+    "returncode": 0,
+    "stdout_tail": "\n    },\n    {\n      \"extraction_job_id\": 4,\n      \"model_name\": \"muq\",\n      \"model_version\": \"large-msd-iter\",\n      \"embedding_dim\": 768,\n      \"target_scope\": \"reference_set:phase1_hot_reference_v1\",\n      \"required_packages\": [\n        \"numpy\",\n        \"torch\",\n        \"torchaudio\",\n        \"transformers\"\n      ],\n      \"missing_packages\": [\n        \"torch\",\n        \"torchaudio\",\n        \"transformers\"\n      ],\n      \"downloads_root_exists\": false,\n      \"ready_for_live_worker\": false\n    },\n    {\n      \"extraction_job_id\": 5,\n      \"model_name\": \"ecapa\",\n      \"model_version\": \"acr-baseline-v1\",\n      \"embedding_dim\": 192,\n      \"target_scope\": \"reference_set:phase1_hot_reference_v1\",\n      \"required_packages\": [\n        \"numpy\",\n        \"torch\",\n        \"torchaudio\",\n        \"speechbrain\"\n      ],\n      \"missing_packages\": [\n        \"torch\",\n        \"torchaudio\",\n        \"speechbrain\"\n      ],\n      \"downloads_root_exists\": false,\n      \"ready_for_live_worker\": false\n    }\n  ],\n  \"summary\": {\n    \"total_jobs\": 5,\n    \"ready_jobs\": 0,\n    \"blocked_jobs\": 5,\n    \"missing_packages_union\": [\n      \"speechbrain\",\n      \"torch\",\n      \"torchaudio\",\n      \"transformers\"\n    ]\n  }\n}\n",
+    "stderr_tail": "",
+    "passed": true
+  },
+  "worker_contract_smoke": {
+    "command": "cd /workspace/acr-engine && PG_DSN=\"${PG_DSN:?set PG_DSN}\" /usr/local/miniconda3/bin/python scripts/run_phase1_worker_contract_smoke_live.py --dsn \"$PG_DSN\" --schema acr_test --output data/pgvector_eval/music20/phase1_worker_contract_smoke_report.json",
+    "returncode": 0,
+    "stdout_tail": "{\n  \"schema\": \"acr_test\",\n  \"dsn_redacted\": \"postgres://d2:***@127.0.0.1:5432/d2\",\n  \"exact_lane\": {\n    \"job_id\": 1,\n    \"returncode\": 0,\n    \"job_status\": \"failed\",\n    \"failure_reason\": \"unreadable_audio_assets\",\n    \"missing_asset_count\": 20,\n    \"artifact\": \"data/pgvector_eval/music20/phase1_worker_contract_smoke_exact.json\"\n  },\n  \"semantic_lane\": {\n    \"returncode\": 0,\n    \"semantic_job_count\": 4,\n    \"failed_jobs\": 4,\n    \"unique_blockers\": [\n      \"model_runtime_unavailable\",\n      \"unreadable_audio_assets\"\n    ],\n    \"artifact\": \"data/pgvector_eval/music20/phase1_worker_contract_smoke_semantic_matrix.json\"\n  },\n  \"summary\": {\n    \"exact_status\": \"failed\",\n    \"semantic_failed_jobs\": 4,\n    \"shared_environment_blockers\": [\n      \"missing /workspace/downloads mount\",\n      \"missing semantic model runtime dependencies\"\n    ]\n  }\n}\n",
+    "stderr_tail": "",
+    "passed": true
+  }
+}
\ No newline at end of file

 ## 2026-06-04
 ## 2026-06-04
 
 
+- 新增 `phase1_validation_commands_execution_report.json`，直接从 `phase1_extraction_plan_report.json` 读取并执行 `validation_commands.prereq_audit` 与 `validation_commands.worker_contract_smoke`，两条命令均返回 `0`，证明 planner 产物可被脚本化直接消费。
 - 更新 `scripts/plan_phase1_extraction_jobs_live.py` 与 `phase1_extraction_plan_report.json`，除了 per-job `command_suggestions` 之外，又补充了 `validation_commands`：`prereq_audit`、`worker_contract_smoke`、`semantic_vector_negative_matrix`、`asset_level_upsert_validation`，使 planner 本身也成为下次 session 的执行入口。
 - 更新 `scripts/plan_phase1_extraction_jobs_live.py` 与 `phase1_extraction_plan_report.json`，除了 per-job `command_suggestions` 之外，又补充了 `validation_commands`：`prereq_audit`、`worker_contract_smoke`、`semantic_vector_negative_matrix`、`asset_level_upsert_validation`，使 planner 本身也成为下次 session 的执行入口。
 - 新增 `scripts/run_phase1_prereq_audit_live.py` 与 `phase1_prereq_audit_report.json`，把 `/workspace/downloads` 挂载状态、`torch/torchaudio/transformers/speechbrain` 依赖状态与 5 条 Phase-1 jobs 的 readiness 汇总到一份 live 审计报告；当前结果为 `ready_jobs=0`、`blocked_jobs=5`。
 - 新增 `scripts/run_phase1_prereq_audit_live.py` 与 `phase1_prereq_audit_report.json`，把 `/workspace/downloads` 挂载状态、`torch/torchaudio/transformers/speechbrain` 依赖状态与 5 条 Phase-1 jobs 的 readiness 汇总到一份 live 审计报告；当前结果为 `ready_jobs=0`、`blocked_jobs=5`。
 - 新增 `scripts/run_embedding_vector_table_negative_matrix_live.py` 与 `embedding_vector_table_negative_matrix_report.json`，在 live PostgreSQL 上补齐 semantic preflight 的三类向量表负例：维度不匹配、未 allowlist、schema 缺表；三类 case 都会稳定落到 `preflight_failed`，且 `vector_table_report.reason` 与预期一致。
 - 新增 `scripts/run_embedding_vector_table_negative_matrix_live.py` 与 `embedding_vector_table_negative_matrix_report.json`，在 live PostgreSQL 上补齐 semantic preflight 的三类向量表负例：维度不匹配、未 allowlist、schema 缺表；三类 case 都会稳定落到 `preflight_failed`，且 `vector_table_report.reason` 与预期一致。

@@ -468,3 +468,19 @@ cd /workspace/acr-engine && PG_DSN="${PG_DSN:?set PG_DSN}" EXTRACTION_JOB_ID=2 F
 4. 再验证 asset-level upsert contract
 4. 再验证 asset-level upsert contract
 
 
 这让 planner 从“只会排任务”升级成“同时给出执行前检查入口”的交付物。
 这让 planner 从“只会排任务”升级成“同时给出执行前检查入口”的交付物。
+
+
+### 10.5 planner validation_commands 已做直接消费验证
+
+本轮继续补了一层 fresh evidence：不是只看 planner JSON 里有命令，而是**直接从 `phase1_extraction_plan_report.json` 读取命令后执行**。
+
+对应产物：
+
+- `acr-engine/data/pgvector_eval/music20/phase1_validation_commands_execution_report.json`
+
+当前已验证：
+
+- `validation_commands.prereq_audit` -> `returncode = 0`
+- `validation_commands.worker_contract_smoke` -> `returncode = 0`
+
+这说明 planner 报告现在不仅能“展示命令”，还可以被脚本化消费为真正的执行入口。

@@ -197,6 +197,7 @@ sed -n '1,320p' acr-engine/sql/acr_pg_schema_v2.sql
 - `scripts/run_embedding_vector_table_negative_matrix_live.py` 已在 live PostgreSQL 上补齐 semantic vector-table 负例矩阵：`vector_table_dim_mismatch`、`vector_table_not_allowlisted`、`vector_table_missing_in_schema` 三类错误都能被稳定写入 `vector_table_report.reason`
 - `scripts/run_embedding_vector_table_negative_matrix_live.py` 已在 live PostgreSQL 上补齐 semantic vector-table 负例矩阵：`vector_table_dim_mismatch`、`vector_table_not_allowlisted`、`vector_table_missing_in_schema` 三类错误都能被稳定写入 `vector_table_report.reason`
 - `scripts/run_phase1_prereq_audit_live.py` 已给出当前 host 的先决条件审计：`downloads_root_exists=false`、`ready_jobs=0/5`，并把 `torch/torchaudio/transformers/speechbrain` 的缺失状态按 job 落成 JSON 报告
 - `scripts/run_phase1_prereq_audit_live.py` 已给出当前 host 的先决条件审计：`downloads_root_exists=false`、`ready_jobs=0/5`，并把 `torch/torchaudio/transformers/speechbrain` 的缺失状态按 job 落成 JSON 报告
 - `phase1_extraction_plan_report.json` 现已附带 `validation_commands`，下次 session 可以直接从 planner 复制 `prereq_audit / worker_contract_smoke / semantic_vector_negative_matrix / asset_level_upsert_validation` 四类命令
 - `phase1_extraction_plan_report.json` 现已附带 `validation_commands`，下次 session 可以直接从 planner 复制 `prereq_audit / worker_contract_smoke / semantic_vector_negative_matrix / asset_level_upsert_validation` 四类命令
+- `phase1_validation_commands_execution_report.json` 已证明 planner 里的 `prereq_audit` 与 `worker_contract_smoke` 两条 validation commands 可以被直接脚本消费且 `returncode=0`
 - `phase1_hot_reference_v1` 在 `acr_test` 里已经真实补齐 `20` 个 reference members，因此 worker dry-run 当前看到的 scope 已是 `20 recordings / 20 assets / 20 windows`
 - `phase1_hot_reference_v1` 在 `acr_test` 里已经真实补齐 `20` 个 reference members，因此 worker dry-run 当前看到的 scope 已是 `20 recordings / 20 assets / 20 windows`
 - worker contract 现在已有基础前置状态保护；重复执行同一 chromaprint dry-run job 会被 `expected_status=pending` 明确拒绝，证据见 `phase1_worker_double_claim_guard_report.json`
 - worker contract 现在已有基础前置状态保护；重复执行同一 chromaprint dry-run job 会被 `expected_status=pending` 明确拒绝，证据见 `phase1_worker_double_claim_guard_report.json`
 - exact lane 的 `run_chromaprint_job.py` 已具备非 dry-run 写入路径；当前在 `acr_test` 的 live 结果是因为 `/workspace/downloads/...` 缺失而明确 `failed`，不是继续假装 `completed`
 - exact lane 的 `run_chromaprint_job.py` 已具备非 dry-run 写入路径；当前在 `acr_test` 的 live 结果是因为 `/workspace/downloads/...` 缺失而明确 `failed`，不是继续假装 `completed`