core-runtime 2.12.0__tar.gz → 2.13.0__tar.gz

{core_runtime-2.12.0 → core_runtime-2.13.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: core-runtime
-Version: 2.12.0
+Version: 2.13.0
 Summary: CORE: executable constitutional governance for AI-assisted software development.
 License-Expression: MIT
 License-File: LICENSE
@@ -18,7 +18,6 @@ Classifier: Programming Language :: Python :: 3.12
 Classifier: Topic :: Software Development :: Quality Assurance
 Classifier: Topic :: Software Development :: Testing
 Classifier: Topic :: System :: Systems Administration
-Provides-Extra: telemetry
 Requires-Dist: PyJWT (>=2.15.1,<3.0)
 Requires-Dist: argon2-cffi (>=23.1,<24.0)
 Requires-Dist: asyncpg (>=0.31.0,<0.32.0)
@@ -31,12 +30,9 @@ Requires-Dist: fastapi (>=0.136.0,<0.137.0)
 Requires-Dist: filelock (>=3.32.7,<4.0.0)
 Requires-Dist: httpx (>=0.28.0,<0.29.0)
 Requires-Dist: jsonschema (>=4.26.0,<5)
+Requires-Dist: mcp (>=2.3.0,<2.4.0)
 Requires-Dist: msgpack (>=1.2.3,<2.0.0)
 Requires-Dist: numpy (>=2.5.3,<3.0.0)
-Requires-Dist: opentelemetry-api (>=1.21.0,<2.0.0) ; extra == "telemetry"
-Requires-Dist: opentelemetry-instrumentation-fastapi (>=0.42b0,<0.43) ; extra == "telemetry"
-Requires-Dist: opentelemetry-instrumentation-sqlalchemy (>=0.42b0,<0.43) ; extra == "telemetry"
-Requires-Dist: opentelemetry-sdk (>=1.21.0,<2.0.0) ; extra == "telemetry"
 Requires-Dist: pathspec (>=1.1.0,<2.0.0)
 Requires-Dist: pydantic (>=2.13.5,<3.0)
 Requires-Dist: pydantic-settings (>=2.15.0,<3.0.0)
@@ -64,7 +60,7 @@ Description-Content-Type: text/markdown
 > Designed for environments where AI action traceability is not optional.
 
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
-[![Release](https://img.shields.io/badge/Release-v2.12.0-blue)](https://github.com/DariuszNewecki/CORE/releases)
+[![Release](https://img.shields.io/badge/Release-v2.13.0-blue)](https://github.com/DariuszNewecki/CORE/releases)
 [![Docs](https://img.shields.io/badge/Docs-online-green)](https://dariusznewecki.github.io/CORE/)
 [![Autonomy](https://img.shields.io/badge/Autonomy-A3%20unproven-orange)](https://github.com/DariuszNewecki/CORE/blob/main/.specs/attestations/production-readiness.yaml)
 
@@ -237,7 +233,7 @@ Every autonomous operation is governed by the same constitutional loop:
 ```mermaid
 flowchart TD
     A["🟢 GOAL\nHUMAN INTENT"] --> B["📂 CONTEXT\nRepo state • knowledge • history"]
-    B --> C["🔒 CONSTRAINTS\nImmutable rules\n264 rules • 16 engines"]
+    B --> C["🔒 CONSTRAINTS\nImmutable rules\n265 rules • 17 engines"]
     C --> D["🗺️ PLAN\nStep-by-step reasoning\nRule-aware plan"]
     D --> E["✨ GENERATE\nCode • changes • tool calls"]
     E --> F["✅ VALIDATE\nDeterministic checks\nAST • semantic • intent • style"]
@@ -325,6 +321,8 @@ Enforcement strengths: **Blocking** · **Reporting** · **Advisory**
 | `taxonomy_gate`   | Capability-id ↔ atomic-action coherence (ADR-079 D9) |
 | `contracts_gate`  | Cross-cutting data-contract coherence (context-level; ADR-102) |
 | `runtime_gate`    | Runtime telemetry checks over blackboard data, not source (ADR-081/082) |
+| `reference_gate`  | Concrete `.intent/` references resolve (context-level) |
+| `exclusion_gate`  | Every `scope.excludes` entry still exempts something (context-level) |
 | `attestation_gate`| The honest third outcome: requirements no automated method can settle need a dated human attestation (ADR-113) |
 | `llm_gate`        | LLM-assisted semantic checks                 |
 | `grc_judge`       | LLM-assisted compliance assessment of GRC documents |
@@ -334,7 +332,7 @@ Enforcement strengths: **Blocking** · **Reporting** · **Advisory**
 
 Deterministic when possible. LLM only when necessary.
 
-264 rules across 61 rule documents. 254 are mapped to enforcement engines; the other 10 are advisory rules that are mapping-exempt by design (#820), so every rule that requires a mapping has one. "Mapped" means engine-bound — not enforced in every mode: stateless CI skips `knowledge_gate` and `llm_gate`, which need the knowledge graph and an LLM provider.
+265 rules across 61 rule documents. 255 are mapped to enforcement engines; the other 10 are advisory rules that are mapping-exempt by design (#820), so every rule that requires a mapping has one. "Mapped" means engine-bound — not enforced in every mode: stateless CI skips `knowledge_gate` and `llm_gate`, which need the knowledge graph and an LLM provider.
 
 ---
 
@@ -450,7 +448,7 @@ CORE's tracker mixes governance-internal bookkeeping with ordinary engineering w
 
 ## Project Status
 
-**Current Release:** v2.12.0 — CORE knows its own state: every verdict names the law it evaluated, an append-only state ledger, a stale daemon stops acting; verdicts mean the same for every project; PyPI publishes only from a green-CI commit
+**Current Release:** v2.13.0 — your AI assistant can ask CORE instead of guessing: law, decisions and a change verdict over MCP, verified end to end on a clean Ubuntu 26.04 LTS machine
 
 Active work: A3 Governed Autonomy. The mechanism is built — the daemon runs continuously,
 finds constitutional violations in its own codebase, proposes fixes, routes them through the

{core_runtime-2.12.0 → core_runtime-2.13.0}/README.md

@@ -5,7 +5,7 @@
 > Designed for environments where AI action traceability is not optional.
 
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
-[![Release](https://img.shields.io/badge/Release-v2.12.0-blue)](https://github.com/DariuszNewecki/CORE/releases)
+[![Release](https://img.shields.io/badge/Release-v2.13.0-blue)](https://github.com/DariuszNewecki/CORE/releases)
 [![Docs](https://img.shields.io/badge/Docs-online-green)](https://dariusznewecki.github.io/CORE/)
 [![Autonomy](https://img.shields.io/badge/Autonomy-A3%20unproven-orange)](https://github.com/DariuszNewecki/CORE/blob/main/.specs/attestations/production-readiness.yaml)
 
@@ -178,7 +178,7 @@ Every autonomous operation is governed by the same constitutional loop:
 ```mermaid
 flowchart TD
     A["🟢 GOAL\nHUMAN INTENT"] --> B["📂 CONTEXT\nRepo state • knowledge • history"]
-    B --> C["🔒 CONSTRAINTS\nImmutable rules\n264 rules • 16 engines"]
+    B --> C["🔒 CONSTRAINTS\nImmutable rules\n265 rules • 17 engines"]
     C --> D["🗺️ PLAN\nStep-by-step reasoning\nRule-aware plan"]
     D --> E["✨ GENERATE\nCode • changes • tool calls"]
     E --> F["✅ VALIDATE\nDeterministic checks\nAST • semantic • intent • style"]
@@ -266,6 +266,8 @@ Enforcement strengths: **Blocking** · **Reporting** · **Advisory**
 | `taxonomy_gate`   | Capability-id ↔ atomic-action coherence (ADR-079 D9) |
 | `contracts_gate`  | Cross-cutting data-contract coherence (context-level; ADR-102) |
 | `runtime_gate`    | Runtime telemetry checks over blackboard data, not source (ADR-081/082) |
+| `reference_gate`  | Concrete `.intent/` references resolve (context-level) |
+| `exclusion_gate`  | Every `scope.excludes` entry still exempts something (context-level) |
 | `attestation_gate`| The honest third outcome: requirements no automated method can settle need a dated human attestation (ADR-113) |
 | `llm_gate`        | LLM-assisted semantic checks                 |
 | `grc_judge`       | LLM-assisted compliance assessment of GRC documents |
@@ -275,7 +277,7 @@ Enforcement strengths: **Blocking** · **Reporting** · **Advisory**
 
 Deterministic when possible. LLM only when necessary.
 
-264 rules across 61 rule documents. 254 are mapped to enforcement engines; the other 10 are advisory rules that are mapping-exempt by design (#820), so every rule that requires a mapping has one. "Mapped" means engine-bound — not enforced in every mode: stateless CI skips `knowledge_gate` and `llm_gate`, which need the knowledge graph and an LLM provider.
+265 rules across 61 rule documents. 255 are mapped to enforcement engines; the other 10 are advisory rules that are mapping-exempt by design (#820), so every rule that requires a mapping has one. "Mapped" means engine-bound — not enforced in every mode: stateless CI skips `knowledge_gate` and `llm_gate`, which need the knowledge graph and an LLM provider.
 
 ---
 
@@ -391,7 +393,7 @@ CORE's tracker mixes governance-internal bookkeeping with ordinary engineering w
 
 ## Project Status
 
-**Current Release:** v2.12.0 — CORE knows its own state: every verdict names the law it evaluated, an append-only state ledger, a stale daemon stops acting; verdicts mean the same for every project; PyPI publishes only from a green-CI commit
+**Current Release:** v2.13.0 — your AI assistant can ask CORE instead of guessing: law, decisions and a change verdict over MCP, verified end to end on a clean Ubuntu 26.04 LTS machine
 
 Active work: A3 Governed Autonomy. The mechanism is built — the daemon runs continuously,
 finds constitutional violations in its own codebase, proposes fixes, routes them through the

{core_runtime-2.12.0 → core_runtime-2.13.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
 
 [project]
 name = "core-runtime"
-version = "2.12.0"
+version = "2.13.0"
 description = "CORE: executable constitutional governance for AI-assisted software development."
 authors = [
     { name = "Dariusz Newecki", email = "d.newecki@gmail.com" },
@@ -74,6 +74,8 @@ dependencies = [
     "pathspec>=1.1.0,<2.0.0",
     # Dependency Injection
     "dependency-injector>=4.49.1,<5.0.0",
+    # Assistant surface: the MCP server (ADR-168 D4.3)
+    "mcp>=2.3.0,<2.4.0",
     # Rate limiting (optional Redis backend)
     "redis>=5.0,<6.0",
     # Security floors for transitive deps (pip-audit; not directly imported):
@@ -83,15 +85,6 @@ dependencies = [
     "setuptools>=84.0.0",
 ]
 
-[project.optional-dependencies]
-# Observability (Optional) — install via `pip install core-runtime[telemetry]`
-telemetry = [
-    "opentelemetry-api>=1.21.0,<2.0.0",
-    "opentelemetry-sdk>=1.21.0,<2.0.0",
-    "opentelemetry-instrumentation-fastapi>=0.42b0,<0.43",
-    "opentelemetry-instrumentation-sqlalchemy>=0.42b0,<0.43",
-]
-
 [project.urls]
 Homepage = "https://github.com/DariuszNewecki/CORE"
 Repository = "https://github.com/DariuszNewecki/CORE"

{core_runtime-2.12.0 → core_runtime-2.13.0}/src/api/v1/secrets_routes.py

@@ -153,7 +153,7 @@ async def delete_secret(
 ) -> dict:
     """Permanently delete an encrypted secret."""
     try:
-        await svc.delete_secret(session, key)
+        await svc.delete_secret(session, key, audit_context="api:delete")
     except SecretNotFoundError:
         raise HTTPException(status_code=404, detail=f"Secret '{key}' not found.")
     return {"key": key, "deleted": True}

core_runtime-2.13.0/src/body/atomic/action_audit.py

@@ -0,0 +1,121 @@
+# src/body/atomic/action_audit.py
+"""
+Action audit-trail persistence for the ActionExecutor gateway.
+
+Moved verbatim out of ``ActionExecutor._audit_log`` (ADR-095 split of the
+too-large executor class; zero behavior change). ``ActionAuditRecorder``
+writes one ``core.action_results`` row per executed action — step 7 of
+``ActionExecutor.execute`` — with bounded retry/backoff, and surfaces an
+``AUDIT_GAP`` error for write actions whose row could not be persisted.
+
+This module defines no atomic action and never invokes ``ActionExecutor``;
+it is a plain persistence helper the executor delegates to.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+from typing import TYPE_CHECKING
+
+from sqlalchemy import text
+
+from shared.logger import _current_run_id, getLogger
+
+
+if TYPE_CHECKING:
+    from body.atomic.registry import ActionDefinition
+    from shared.action_types import ActionResult
+    from shared.context import CoreContext
+
+logger = getLogger(__name__)
+
+
+# ID: 7c763fc4-5116-4ad9-a05d-761811083559
+class ActionAuditRecorder:
+    """Persists ActionExecutor results to the core.action_results audit trail."""
+
+    _AUDIT_MAX_ATTEMPTS: int = 3
+    _AUDIT_BACKOFF_BASE_SEC: float = 0.1  # doubled each retry: 0.1s, 0.2s
+
+    def __init__(self, core_context: CoreContext):
+        self.core_context = core_context
+
+    # ID: 454c8ccb-ece8-4ef8-baf1-13c9c19f4300
+    async def record(
+        self, definition: ActionDefinition, result: ActionResult, write: bool
+    ) -> None:
+        """
+        Log action execution to database audit trail (SSOT).
+
+        Retries the INSERT up to _AUDIT_MAX_ATTEMPTS times with exponential
+        backoff before declaring an AUDIT_GAP. Each attempt opens a fresh
+        session so a rolled-back transaction from a prior attempt does not
+        poison the retry.
+
+        CONSTITUTIONAL FIX: session_id is read cleanly from _current_run_id
+        context var (imported at module level). Removed duplicate key and
+        broken __import__ hack from prior patch.
+        """
+        stmt = text(
+            """
+            INSERT INTO core.action_results
+            (action_type, ok, file_path, error_message, action_metadata, agent_id, duration_ms)
+            VALUES (:atype, :ok, :path, :err, :meta, :agent, :dur)
+            """
+        )
+        # Prefer session_id from core_context, fall back to context var
+        session_id = getattr(
+            self.core_context, "session_id", None
+        ) or _current_run_id.get(None)
+        params = {
+            "atype": definition.action_id,
+            "ok": result.ok,
+            "path": result.data.get("path") or result.data.get("file_path"),
+            "err": result.data.get("error") if not result.ok else None,
+            "meta": json.dumps(
+                {
+                    "write_mode": write,
+                    "impact": definition.impact_level,
+                    "session_id": session_id,
+                }
+            ),
+            "agent": "ActionExecutor",
+            "dur": int(result.duration_sec * 1000),
+        }
+
+        last_exc: Exception | None = None
+        for attempt in range(1, self._AUDIT_MAX_ATTEMPTS + 1):
+            try:
+                async with self.core_context.registry.session() as session:
+                    async with session.begin():
+                        await session.execute(stmt, params)
+                return  # persisted successfully
+            except Exception as e:
+                last_exc = e
+                if attempt < self._AUDIT_MAX_ATTEMPTS:
+                    await asyncio.sleep(
+                        self._AUDIT_BACKOFF_BASE_SEC * (2 ** (attempt - 1))
+                    )
+
+        # All attempts exhausted — #634/#752: surface LOUD for write actions.
+        # Audit persistence runs at step 7 after the mutation has already
+        # landed; there is no file+DB transaction to unwind. On DB
+        # unavailability/serialization failure only (schema has no per-row
+        # failure mode — action_type/ok are always supplied).
+        assert last_exc is not None
+        if write:
+            logger.error(
+                "AUDIT_GAP: write action %s executed but its "
+                "core.action_results row failed to persist after %d attempts "
+                "(%s) — mutation stands, audit trail incomplete (#634/#752)",
+                definition.action_id,
+                self._AUDIT_MAX_ATTEMPTS,
+                last_exc,
+            )
+        else:
+            logger.warning(
+                "Non-blocking audit log failure (read) after %d attempts: %s",
+                self._AUDIT_MAX_ATTEMPTS,
+                last_exc,
+            )

{core_runtime-2.12.0 → core_runtime-2.13.0}/src/body/atomic/check_actions.py

@@ -17,9 +17,7 @@ Enforces:
 
 from __future__ import annotations
 
-import asyncio
 import json
-import subprocess
 import time
 from pathlib import Path
 from typing import TYPE_CHECKING, Any
@@ -28,6 +26,7 @@ from body.atomic.registry import ActionCategory, register_action
 from shared.action_types import ActionImpact, ActionResult
 from shared.atomic_action import atomic_action
 from shared.logger import getLogger
+from shared.utils.subprocess_utils import run_command_async
 
 
 if TYPE_CHECKING:
@@ -105,13 +104,7 @@ async def action_check_imports(
     ]
 
     try:
-        result = await asyncio.to_thread(
-            subprocess.run,
-            cmd,
-            capture_output=True,
-            text=True,
-            check=False,
-        )
+        result = await run_command_async(cmd)
 
         violations: list[dict] = []
 

{core_runtime-2.12.0 → core_runtime-2.13.0}/src/body/atomic/executor.py

@@ -18,15 +18,12 @@ CRITICAL: This enforces the "single execution contract" principle.
 
 from __future__ import annotations
 
-import asyncio
 import inspect
-import json
 import time
 from pathlib import Path
 from typing import TYPE_CHECKING, Any
 
-from sqlalchemy import text
-
+from body.atomic.action_audit import ActionAuditRecorder
 from body.atomic.registry import ActionCategory, ActionDefinition, action_registry
 from body.atomic.sandbox_lifecycle import SandboxLifecycle
 from shared.action_types import ActionImpact, ActionResult
@@ -38,7 +35,7 @@ from shared.infrastructure.intent.action_risk import (
     load_safe_auto_approval_envelope,
 )
 from shared.infrastructure.intent.intent_repository import get_intent_repository
-from shared.logger import _current_run_id, getLogger
+from shared.logger import getLogger
 
 
 if TYPE_CHECKING:
@@ -481,8 +478,8 @@ class ActionExecutor:
         # 6. Post-execution hooks
         await self._post_execute_hooks(definition, result)
 
-        # 7. Audit logging
-        await self._audit_log(definition, result, write)
+        # 7. Audit logging (persistence lives in body.atomic.action_audit)
+        await ActionAuditRecorder(self.core_context).record(definition, result, write)
 
         return result
 
@@ -589,88 +586,6 @@ class ActionExecutor:
         """
         logger.debug("Post-execution hooks for %s", definition.action_id)
 
-    _AUDIT_MAX_ATTEMPTS: int = 3
-    _AUDIT_BACKOFF_BASE_SEC: float = 0.1  # doubled each retry: 0.1s, 0.2s
-
-    # ID: 454c8ccb-ece8-4ef8-baf1-13c9c19f4300
-    async def _audit_log(
-        self, definition: ActionDefinition, result: ActionResult, write: bool
-    ) -> None:
-        """
-        Log action execution to database audit trail (SSOT).
-
-        Retries the INSERT up to _AUDIT_MAX_ATTEMPTS times with exponential
-        backoff before declaring an AUDIT_GAP. Each attempt opens a fresh
-        session so a rolled-back transaction from a prior attempt does not
-        poison the retry.
-
-        CONSTITUTIONAL FIX: session_id is read cleanly from _current_run_id
-        context var (imported at module level). Removed duplicate key and
-        broken __import__ hack from prior patch.
-        """
-        stmt = text(
-            """
-            INSERT INTO core.action_results
-            (action_type, ok, file_path, error_message, action_metadata, agent_id, duration_ms)
-            VALUES (:atype, :ok, :path, :err, :meta, :agent, :dur)
-            """
-        )
-        # Prefer session_id from core_context, fall back to context var
-        session_id = getattr(
-            self.core_context, "session_id", None
-        ) or _current_run_id.get(None)
-        params = {
-            "atype": definition.action_id,
-            "ok": result.ok,
-            "path": result.data.get("path") or result.data.get("file_path"),
-            "err": result.data.get("error") if not result.ok else None,
-            "meta": json.dumps(
-                {
-                    "write_mode": write,
-                    "impact": definition.impact_level,
-                    "session_id": session_id,
-                }
-            ),
-            "agent": "ActionExecutor",
-            "dur": int(result.duration_sec * 1000),
-        }
-
-        last_exc: Exception | None = None
-        for attempt in range(1, self._AUDIT_MAX_ATTEMPTS + 1):
-            try:
-                async with self.core_context.registry.session() as session:
-                    async with session.begin():
-                        await session.execute(stmt, params)
-                return  # persisted successfully
-            except Exception as e:
-                last_exc = e
-                if attempt < self._AUDIT_MAX_ATTEMPTS:
-                    await asyncio.sleep(
-                        self._AUDIT_BACKOFF_BASE_SEC * (2 ** (attempt - 1))
-                    )
-
-        # All attempts exhausted — #634/#752: surface LOUD for write actions.
-        # Audit persistence runs at step 7 after the mutation has already
-        # landed; there is no file+DB transaction to unwind. On DB
-        # unavailability/serialization failure only (schema has no per-row
-        # failure mode — action_type/ok are always supplied).
-        assert last_exc is not None
-        if write:
-            logger.error(
-                "AUDIT_GAP: write action %s executed but its "
-                "core.action_results row failed to persist after %d attempts "
-                "(%s) — mutation stands, audit trail incomplete (#634/#752)",
-                definition.action_id,
-                self._AUDIT_MAX_ATTEMPTS,
-                last_exc,
-            )
-        else:
-            logger.warning(
-                "Non-blocking audit log failure (read) after %d attempts: %s",
-                self._AUDIT_MAX_ATTEMPTS,
-                last_exc,
-            )
-
     # ID: eff3eded-b30d-49e0-b50c-3503a1b695af
     def _prepare_params(
         self,

{core_runtime-2.12.0 → core_runtime-2.13.0}/src/body/atomic/modularity_fix.py

@@ -366,7 +366,7 @@ async def action_fix_modularity(
         )
         # Delete original monolith — the package __init__.py re-exports
         # all symbols so callers resolve without import-path changes.
-        # Mirrors _execute_deterministic_split in the workflow path.
+        # Mirrors DeterministicSplitPlanner.execute in the workflow path.
         if target.exists():
             core_context.file_handler.remove_file(rel_path)
             logger.info("fix.modularity: deleted original monolith %s", rel_path)

{core_runtime-2.12.0 → core_runtime-2.13.0}/src/body/maintenance/scripts/context_export.py

@@ -20,12 +20,14 @@ import urllib.request
 from pathlib import Path
 from typing import TYPE_CHECKING
 
+from body.infrastructure.storage.file_handler import FileHandler
 from shared.infrastructure.git_service import GitService
 from shared.infrastructure.intent.operational_config import load_operational_config
-from body.infrastructure.storage.file_handler import FileHandler
 from shared.logger import getLogger
 from shared.path_resolver import PathResolver
 from shared.time import now_iso
+from shared.utils.subprocess_utils import run_command_async
+
 
 if TYPE_CHECKING:
     from shared.context import CoreContext
@@ -93,7 +95,9 @@ class ContextExporter:
         _exports = PathResolver(self.repo_root).exports_dir
         self.output_base = output_base or _exports
         self.timestamp = now_iso().replace(":", "-").split(".")[0]
-        self.export_rel_dir = str(_exports.relative_to(self.repo_root)) + f"/core_export_{self.timestamp}"
+        self.export_rel_dir = (
+            str(_exports.relative_to(self.repo_root)) + f"/core_export_{self.timestamp}"
+        )
 
         # Mutation surface
         self.fh = FileHandler(str(self.repo_root))
@@ -191,24 +195,19 @@ class ContextExporter:
         )
 
     async def _export_db_schema(self):
-        """Capture DB schema using subprocess, persisted via FileHandler."""
+        """Capture DB schema via pg_dump (subprocess_utils), persisted via FileHandler."""
         logger.info("🗄️ Capturing Database Schema...")
         db_url = self._context.settings.DATABASE_URL
 
         try:
             # Note: requires pg_dump installed on host
-            proc = await asyncio.create_subprocess_exec(
-                "pg_dump",
-                "--schema-only",
-                "--no-owner",
-                db_url,
-                stdout=asyncio.subprocess.PIPE,
-                stderr=asyncio.subprocess.PIPE,
+            result = await run_command_async(
+                ["pg_dump", "--schema-only", "--no-owner", db_url]
             )
-            stdout, _ = await proc.communicate()
-            if stdout:
+            if result.stdout:
+                # run_command_async strips output; restore the trailing newline.
                 self.fh.write_runtime_text(
-                    f"{self.export_rel_dir}/db_schema.sql", stdout.decode()
+                    f"{self.export_rel_dir}/db_schema.sql", result.stdout + "\n"
                 )
         except Exception as e:
             logger.warning("Could not export DB schema: %s", e)

{core_runtime-2.12.0 → core_runtime-2.13.0}/src/body/project_lifecycle/bootstrap_service.py

@@ -8,10 +8,10 @@ such as creating a default set of GitHub issues for a new repository.
 from __future__ import annotations
 
 import shutil
-import subprocess
 
 from shared.exceptions import CoreError
 from shared.logger import getLogger
+from shared.utils.subprocess_utils import run_command
 
 
 logger = getLogger(__name__)
@@ -87,12 +87,10 @@ def _run_gh_command(command: list[str], ignore_errors: bool = False):
         logger.info("Install GitHub CLI to use bootstrap features.")
         raise BootstrapError("'gh' (GitHub CLI) not found in PATH.", exit_code=1)
 
-    try:
-        subprocess.run(command, check=True, capture_output=True, text=True)
-    except subprocess.CalledProcessError as e:
-        if not ignore_errors:
-            logger.error("Error running gh command: %s", e.stderr)
-            raise BootstrapError("Error running gh command.", exit_code=1) from e
+    result = run_command(command)
+    if result.returncode != 0 and not ignore_errors:
+        logger.error("Error running gh command: %s", result.stderr)
+        raise BootstrapError("Error running gh command.", exit_code=1)
 
 
 # ID: 17f7cac0-4134-4885-93fb-0d432c634ed1

{core_runtime-2.12.0 → core_runtime-2.13.0}/src/body/project_lifecycle/integration_service.py

@@ -7,12 +7,11 @@ MOVED: From features/project_lifecycle to body/project_lifecycle (Wave 1 Rebirth
 
 from __future__ import annotations
 
-import asyncio
-
 # REFACTORED: Removed direct settings import
 from shared.context import CoreContext
 from shared.exceptions import CoreError
 from shared.logger import getLogger
+from shared.utils.subprocess_utils import run_command_async
 
 
 logger = getLogger(__name__)
@@ -92,20 +91,16 @@ async def integrate_changes(context: CoreContext, commit_message: str) -> None:
                 step["description"],
             )
             command_parts = step["command"].split()  # type: ignore[attr-defined]
-            process = await asyncio.create_subprocess_exec(
-                *command_parts,
-                stdout=asyncio.subprocess.PIPE,
-                stderr=asyncio.subprocess.PIPE,
-                cwd=context.git_service.repo_path,
+            result = await run_command_async(
+                command_parts, cwd=context.git_service.repo_path
             )
-            stdout, stderr = await process.communicate()
 
-            if stdout:
-                logger.info(stdout.decode())
-            if stderr:
-                logger.warning(stderr.decode())
+            if result.stdout:
+                logger.info(result.stdout)
+            if result.stderr:
+                logger.warning(result.stderr)
 
-            if process.returncode != 0:
+            if result.returncode != 0:
                 if not step.get("continues_on_failure", False):
                     workflow_failed = True
                     break

{core_runtime-2.12.0 → core_runtime-2.13.0}/src/body/quality/coverage_analyzer.py

@@ -11,13 +11,13 @@ from __future__ import annotations
 
 import ast
 import json
-import subprocess
 from pathlib import Path
 from typing import Any
 
 # REFACTORED: Removed direct settings import
 from shared.infrastructure.intent.operational_config import load_operational_config
 from shared.logger import getLogger
+from shared.utils.subprocess_utils import SubprocessTimeoutError, run_command
 
 
 logger = getLogger(__name__)
@@ -43,10 +43,9 @@ class CoverageAnalyzer:
             Dict mapping file paths to coverage percentages
         """
         try:
-            subprocess.run(
+            run_command(
                 ["poetry", "run", "pytest", "--cov=src", "--cov-report=json", "-q"],
                 cwd=self.repo_path,
-                capture_output=True,
                 timeout=_CFG.collect_timeout_sec,
             )
             coverage_json = self.repo_path / "coverage.json"
@@ -117,7 +116,7 @@ class CoverageAnalyzer:
             Dict with coverage metrics or None if measurement fails
         """
         try:
-            result = subprocess.run(
+            result = run_command(
                 [
                     "poetry",
                     "run",
@@ -128,8 +127,6 @@ class CoverageAnalyzer:
                     "-q",
                 ],
                 cwd=self.repo_path,
-                capture_output=True,
-                text=True,
                 timeout=_CFG.full_run_timeout_sec,
             )
             coverage_json = self.repo_path / "coverage.json"
@@ -144,7 +141,7 @@ class CoverageAnalyzer:
                     "timestamp": data.get("meta", {}).get("timestamp"),
                 }
             return self._parse_term_output(result.stdout)
-        except subprocess.TimeoutExpired:
+        except SubprocessTimeoutError:
             logger.error("Coverage measurement timed out after 5 minutes")
             return None
         except Exception as e:

{core_runtime-2.12.0 → core_runtime-2.13.0}/src/body/self_healing/handlers/import_sorting_handler.py

@@ -20,6 +20,7 @@ from body.infrastructure.storage.file_handler import FileHandler
 from body.self_healing.remediation_models import FixResult
 from shared.logger import getLogger
 from shared.models import AuditFinding
+from shared.utils.subprocess_utils import run_command_async
 
 
 logger = getLogger(__name__)
@@ -82,12 +83,7 @@ async def sort_imports_handler(
         cmd = ["ruff", "check", str(file_path), "--select", "I", "--exit-zero"]
         if write:
             cmd.append("--fix")
-        import asyncio
-
-        process = await asyncio.create_subprocess_exec(
-            *cmd, stdout=asyncio.subprocess.PIPE, stderr=asyncio.subprocess.PIPE
-        )
-        _stdout_bytes, _stderr_bytes = await process.communicate()
+        await run_command_async(cmd)
         if write:
             try:
                 new_content = file_path.read_text(encoding="utf-8")