devagent-physical-engine 1.7.1__tar.gz → 1.8.0__tar.gz

{devagent_physical_engine-1.7.1/src/devagent_physical_engine.egg-info → devagent_physical_engine-1.8.0}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: devagent-physical-engine
-Version: 1.7.1
-Summary: Production-grade verification-first robotics evidence runtime with one-command customer intake, validated requirement mapping, adaptive industrial Pilz PTP/LIN report-test Gazebo execution, requirement-driven test generation, selective verification, deterministic Twin lineage, measured physical verification, replay inspection, regression, and FAT evidence.
+Version: 1.8.0
+Summary: Production-grade verification-first robotics evidence runtime with one-command customer intake, truthful robot support tiers, qualified UR5e report-test Gazebo execution, experimental ABB IRB1200 Gazebo/MoveIt/RViz visualization, transfer/place diagnostics, requirement-driven test generation, selective verification, deterministic Twin lineage, measured physical verification, replay inspection, regression, and FAT evidence.
 Author: Tom Ha
 License: All Rights Reserved
 Project-URL: Homepage, https://github.com/tomha85/devagent-physical-engine
@@ -76,7 +76,7 @@ regression + FAT report + Evidence Viewer
 Python 3.11+ is required.
 
 ```bash
-python -m pip install --upgrade devagent-physical-engine==1.7.1
+python -m pip install --upgrade devagent-physical-engine==1.8.0
 ```
 
 Verify the installed package:
@@ -88,7 +88,7 @@ python -c "import devagent_physical_engine as d; print(d.__version__)"
 Expected:
 
 ```text
-1.7.1
+1.8.0
 ```
 
 Run the normal customer workflow:
@@ -115,6 +115,19 @@ For machine-readable CI output:
 devagent verify ./customer-project --json
 ```
 
+Inspect truthful robot support tiers:
+
+```bash
+devagent robots
+```
+
+For the exact ABB IRB1200 experimental visual runtime:
+
+```bash
+devagent robot doctor abb_irb1200
+devagent robot visualize abb_irb1200
+```
+
 Default locations:
 
 ```text
@@ -304,9 +317,52 @@ no fallback to an unqualified simulator or robot
 
 The selected blend radii and every planning attempt receipt remain attached to motion metadata for auditability.
 
+## v1.7.2 transfer/place diagnostic hotfix
+
+A target-stack qualification run proved that `transfer_place` still returned Pilz `FAILURE(99999)` for every v1.7.1 blend candidate, including zero blend. v1.7.2 therefore does not keep guessing at blend values. When that exact bounded failure occurs, DevAgent remains fail-closed and runs two independent **plan-only** probes against the same attached-workpiece PlanningScene:
+
+```text
+lift target
+   ↓
+transfer PTP probe
+   ↓ independent start at transfer target
+place LIN probe
+```
+
+The result is classified deterministically as one of:
+
+```text
+pilz_transfer_segment_planning_failed
+pilz_place_segment_planning_failed
+pilz_transfer_and_place_segment_planning_failed
+pilz_transfer_place_sequence_composition_failed
+```
+
+Each probe writes its own request/receipt plus a `transfer-place-diagnostic.json` summary. These probes never execute motion and never relax targets, planner types, collision policy, attached-workpiece scene state, velocity/acceleration limits, Twin lineage, or real-execution authority. Their purpose is to identify the exact target-stack defect before changing physical execution semantics.
+
+## v1.8 truthful robot support + ABB IRB1200 visual runtime
+
+v1.8 makes robot support explicit instead of implying that every built-in robot profile has the same simulator authority.
+
+Support is exposed through:
+
+```bash
+devagent robots
+devagent robot doctor <robot>
+devagent robot visualize <robot>
+```
+
+The exact `abb_irb1200` profile adds an experimental visual runtime using Ubuntu 24.04, ROS 2 Jazzy, Gazebo Harmonic, `gz_ros2_control`, upstream ABB IRB1200 geometry, MoveIt, and RViz. Runtime readiness is bounded on controller state, `/joint_states`, FollowJointTrajectory, `move_group`, and RViz availability.
+
+For the Golden ABB cell, DevAgent can load the waypoint seed, pre-plan every phase through MoveIt before sending any simulated trajectory, then execute the preview while Gazebo/RViz remain open for inspection.
+
+This is intentionally **visual simulation availability**, not measured FAT qualification. ABB preview evidence is not promoted to a qualified physical measurement, and `real_execution_allowed` remains false. The normal One-Command ABB FAT path continues to fail closed until an exact-Twin measured ABB execution adapter is separately qualified.
+
+Generic `abb_irb`, FANUC CRX, and KUKA KR profiles remain model-only until equivalent vendor/model visual runtimes are implemented and qualified.
+
 ## Qualified visual-execution scope
 
-The packaged visual executor remains intentionally narrow:
+Measured report-test execution remains intentionally narrow:
 
 ```text
 robot:       UR5e
@@ -317,7 +373,7 @@ runtime:     Ubuntu 24.04 + ROS 2 Jazzy + Gazebo Harmonic + MoveIt/Pilz
 
 Unsupported customer Twins, unqualified robot profiles, boundary/fault cases outside the qualified scope, ambiguous case mappings, or incomplete lineage fail closed.
 
-Cross-vendor projects such as ABB may still be imported, normalized, structurally evaluated, and reported, but they do not silently enter the UR5e-only physical executor.
+ABB IRB1200 is separately exposed as `SIMULATION_AVAILABLE_EXPERIMENTAL`: Gazebo/MoveIt/RViz preview is available, but it does not silently enter the UR5e measured FAT executor and cannot promote physical evidence.
 
 ## Interactive engineering session
 
@@ -425,12 +481,12 @@ Ubuntu 24.04
 ROS 2 Jazzy
 Gazebo Harmonic
 gz_ros2_control
-Universal Robots ROS 2 driver
-ur_simulation_gz
-MoveIt 2 + Pilz Industrial Motion Planner
+MoveIt 2
 ```
 
-Preview setup:
+UR5e measured report-test execution additionally uses the Universal Robots ROS 2 driver, `ur_simulation_gz`, and Pilz Industrial Motion Planner.
+
+Preview UR5e setup:
 
 ```bash
 devagent-physical setup --profile ur5e-sim --dry-run
@@ -442,12 +498,24 @@ Apply explicitly:
 devagent-physical setup --profile ur5e-sim --yes
 ```
 
-Check runtime:
+Check the UR5e runtime:
 
 ```bash
 devagent-physical ros doctor
 ```
 
+Check ABB IRB1200 visual runtime prerequisites:
+
+```bash
+devagent robot doctor abb_irb1200
+```
+
+Launch ABB IRB1200 Gazebo + MoveIt + RViz preview:
+
+```bash
+devagent robot visualize abb_irb1200
+```
+
 Hosted CI validates deterministic software contracts. It does not pretend to execute a graphical ROS/Gazebo/MoveIt target stack. Target-stack acceptance must run on the intended workstation.
 
 ## Optional AI providers
@@ -474,7 +542,7 @@ Provider-backed interpretation cannot grant physical truth, site qualification,
 ## Installed CLIs
 
 ```text
-devagent             one-command customer verification + interactive report testing
+devagent             one-command verification, interactive report testing, robot support/visualization
 devagent-commercial  commercial/evidence workflow
 devagent-physical    deterministic core + ROS/qualification tools
 devagent-physical-ai optional provider-backed engineering front end
@@ -492,9 +560,9 @@ devagent-physical-ai optional provider-backed engineering front end
 
 # Project status
 
-**v1.7.1 — Production/Stable software workflow**
+**v1.8.0 — Production/Stable software workflow**
 
-v1.7.1 retains the v1.7 industrial Pilz PTP/LIN execution architecture and adds fail-closed adaptive blend selection for geometry-specific Pilz sequence failures. It also retains the cross-vendor profile gate so non-UR5e customer Twins cannot enter the UR5e-only qualified visual executor.
+v1.8.0 retains the qualified UR5e measured report-test architecture and the v1.7.2 transfer/place diagnostics, while adding truthful robot support tiers and an exact ABB IRB1200 experimental Gazebo Harmonic + MoveIt + RViz visual runtime. ABB visualization is available for engineering preview, but measured FAT promotion remains fail-closed until the exact ABB execution adapter is separately qualified.
 
 Production-oriented software capabilities include:
 
@@ -506,9 +574,12 @@ optional provider-backed advisory criterion proposals
 requirement-driven nominal / boundary / fault case generation
 immutable project/Twin lineage
 exact Requirement → Mapping → Verification Plan → Case provenance
-qualified report-test visual execution contract
+truthful robot support tiers
+qualified UR5e report-test visual execution contract
+experimental exact ABB IRB1200 Gazebo/MoveIt/RViz preview
 industrial Pilz PTP / LIN planning
 adaptive bounded blend selection
+plan-only transfer/place segment diagnostics
 lift-only support-contact ACM lease + exact restore
 Gazebo/MoveIt selected-case visualization
 typed physical measurement binding

{devagent_physical_engine-1.7.1 → devagent_physical_engine-1.8.0}/README.md

@@ -35,7 +35,7 @@ regression + FAT report + Evidence Viewer
 Python 3.11+ is required.
 
 ```bash
-python -m pip install --upgrade devagent-physical-engine==1.7.1
+python -m pip install --upgrade devagent-physical-engine==1.8.0
 ```
 
 Verify the installed package:
@@ -47,7 +47,7 @@ python -c "import devagent_physical_engine as d; print(d.__version__)"
 Expected:
 
 ```text
-1.7.1
+1.8.0
 ```
 
 Run the normal customer workflow:
@@ -74,6 +74,19 @@ For machine-readable CI output:
 devagent verify ./customer-project --json
 ```
 
+Inspect truthful robot support tiers:
+
+```bash
+devagent robots
+```
+
+For the exact ABB IRB1200 experimental visual runtime:
+
+```bash
+devagent robot doctor abb_irb1200
+devagent robot visualize abb_irb1200
+```
+
 Default locations:
 
 ```text
@@ -263,9 +276,52 @@ no fallback to an unqualified simulator or robot
 
 The selected blend radii and every planning attempt receipt remain attached to motion metadata for auditability.
 
+## v1.7.2 transfer/place diagnostic hotfix
+
+A target-stack qualification run proved that `transfer_place` still returned Pilz `FAILURE(99999)` for every v1.7.1 blend candidate, including zero blend. v1.7.2 therefore does not keep guessing at blend values. When that exact bounded failure occurs, DevAgent remains fail-closed and runs two independent **plan-only** probes against the same attached-workpiece PlanningScene:
+
+```text
+lift target
+   ↓
+transfer PTP probe
+   ↓ independent start at transfer target
+place LIN probe
+```
+
+The result is classified deterministically as one of:
+
+```text
+pilz_transfer_segment_planning_failed
+pilz_place_segment_planning_failed
+pilz_transfer_and_place_segment_planning_failed
+pilz_transfer_place_sequence_composition_failed
+```
+
+Each probe writes its own request/receipt plus a `transfer-place-diagnostic.json` summary. These probes never execute motion and never relax targets, planner types, collision policy, attached-workpiece scene state, velocity/acceleration limits, Twin lineage, or real-execution authority. Their purpose is to identify the exact target-stack defect before changing physical execution semantics.
+
+## v1.8 truthful robot support + ABB IRB1200 visual runtime
+
+v1.8 makes robot support explicit instead of implying that every built-in robot profile has the same simulator authority.
+
+Support is exposed through:
+
+```bash
+devagent robots
+devagent robot doctor <robot>
+devagent robot visualize <robot>
+```
+
+The exact `abb_irb1200` profile adds an experimental visual runtime using Ubuntu 24.04, ROS 2 Jazzy, Gazebo Harmonic, `gz_ros2_control`, upstream ABB IRB1200 geometry, MoveIt, and RViz. Runtime readiness is bounded on controller state, `/joint_states`, FollowJointTrajectory, `move_group`, and RViz availability.
+
+For the Golden ABB cell, DevAgent can load the waypoint seed, pre-plan every phase through MoveIt before sending any simulated trajectory, then execute the preview while Gazebo/RViz remain open for inspection.
+
+This is intentionally **visual simulation availability**, not measured FAT qualification. ABB preview evidence is not promoted to a qualified physical measurement, and `real_execution_allowed` remains false. The normal One-Command ABB FAT path continues to fail closed until an exact-Twin measured ABB execution adapter is separately qualified.
+
+Generic `abb_irb`, FANUC CRX, and KUKA KR profiles remain model-only until equivalent vendor/model visual runtimes are implemented and qualified.
+
 ## Qualified visual-execution scope
 
-The packaged visual executor remains intentionally narrow:
+Measured report-test execution remains intentionally narrow:
 
 ```text
 robot:       UR5e
@@ -276,7 +332,7 @@ runtime:     Ubuntu 24.04 + ROS 2 Jazzy + Gazebo Harmonic + MoveIt/Pilz
 
 Unsupported customer Twins, unqualified robot profiles, boundary/fault cases outside the qualified scope, ambiguous case mappings, or incomplete lineage fail closed.
 
-Cross-vendor projects such as ABB may still be imported, normalized, structurally evaluated, and reported, but they do not silently enter the UR5e-only physical executor.
+ABB IRB1200 is separately exposed as `SIMULATION_AVAILABLE_EXPERIMENTAL`: Gazebo/MoveIt/RViz preview is available, but it does not silently enter the UR5e measured FAT executor and cannot promote physical evidence.
 
 ## Interactive engineering session
 
@@ -384,12 +440,12 @@ Ubuntu 24.04
 ROS 2 Jazzy
 Gazebo Harmonic
 gz_ros2_control
-Universal Robots ROS 2 driver
-ur_simulation_gz
-MoveIt 2 + Pilz Industrial Motion Planner
+MoveIt 2
 ```
 
-Preview setup:
+UR5e measured report-test execution additionally uses the Universal Robots ROS 2 driver, `ur_simulation_gz`, and Pilz Industrial Motion Planner.
+
+Preview UR5e setup:
 
 ```bash
 devagent-physical setup --profile ur5e-sim --dry-run
@@ -401,12 +457,24 @@ Apply explicitly:
 devagent-physical setup --profile ur5e-sim --yes
 ```
 
-Check runtime:
+Check the UR5e runtime:
 
 ```bash
 devagent-physical ros doctor
 ```
 
+Check ABB IRB1200 visual runtime prerequisites:
+
+```bash
+devagent robot doctor abb_irb1200
+```
+
+Launch ABB IRB1200 Gazebo + MoveIt + RViz preview:
+
+```bash
+devagent robot visualize abb_irb1200
+```
+
 Hosted CI validates deterministic software contracts. It does not pretend to execute a graphical ROS/Gazebo/MoveIt target stack. Target-stack acceptance must run on the intended workstation.
 
 ## Optional AI providers
@@ -433,7 +501,7 @@ Provider-backed interpretation cannot grant physical truth, site qualification,
 ## Installed CLIs
 
 ```text
-devagent             one-command customer verification + interactive report testing
+devagent             one-command verification, interactive report testing, robot support/visualization
 devagent-commercial  commercial/evidence workflow
 devagent-physical    deterministic core + ROS/qualification tools
 devagent-physical-ai optional provider-backed engineering front end
@@ -451,9 +519,9 @@ devagent-physical-ai optional provider-backed engineering front end
 
 # Project status
 
-**v1.7.1 — Production/Stable software workflow**
+**v1.8.0 — Production/Stable software workflow**
 
-v1.7.1 retains the v1.7 industrial Pilz PTP/LIN execution architecture and adds fail-closed adaptive blend selection for geometry-specific Pilz sequence failures. It also retains the cross-vendor profile gate so non-UR5e customer Twins cannot enter the UR5e-only qualified visual executor.
+v1.8.0 retains the qualified UR5e measured report-test architecture and the v1.7.2 transfer/place diagnostics, while adding truthful robot support tiers and an exact ABB IRB1200 experimental Gazebo Harmonic + MoveIt + RViz visual runtime. ABB visualization is available for engineering preview, but measured FAT promotion remains fail-closed until the exact ABB execution adapter is separately qualified.
 
 Production-oriented software capabilities include:
 
@@ -465,9 +533,12 @@ optional provider-backed advisory criterion proposals
 requirement-driven nominal / boundary / fault case generation
 immutable project/Twin lineage
 exact Requirement → Mapping → Verification Plan → Case provenance
-qualified report-test visual execution contract
+truthful robot support tiers
+qualified UR5e report-test visual execution contract
+experimental exact ABB IRB1200 Gazebo/MoveIt/RViz preview
 industrial Pilz PTP / LIN planning
 adaptive bounded blend selection
+plan-only transfer/place segment diagnostics
 lift-only support-contact ACM lease + exact restore
 Gazebo/MoveIt selected-case visualization
 typed physical measurement binding

{devagent_physical_engine-1.7.1 → devagent_physical_engine-1.8.0}/pyproject.toml

@@ -4,8 +4,8 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "devagent-physical-engine"
-version = "1.7.1"
-description = "Production-grade verification-first robotics evidence runtime with one-command customer intake, validated requirement mapping, adaptive industrial Pilz PTP/LIN report-test Gazebo execution, requirement-driven test generation, selective verification, deterministic Twin lineage, measured physical verification, replay inspection, regression, and FAT evidence."
+version = "1.8.0"
+description = "Production-grade verification-first robotics evidence runtime with one-command customer intake, truthful robot support tiers, qualified UR5e report-test Gazebo execution, experimental ABB IRB1200 Gazebo/MoveIt/RViz visualization, transfer/place diagnostics, requirement-driven test generation, selective verification, deterministic Twin lineage, measured physical verification, replay inspection, regression, and FAT evidence."
 readme = { file = "README.md", content-type = "text/markdown" }
 requires-python = ">=3.11"
 dependencies = ["PyYAML>=6.0,<7"]
@@ -47,7 +47,7 @@ Repository = "https://github.com/tomha85/devagent-physical-engine"
 Issues = "https://github.com/tomha85/devagent-physical-engine/issues"
 
 [project.scripts]
-devagent = "devagent_physical_engine.user_cli_v16:main"
+devagent = "devagent_physical_engine.user_cli_v18:main"
 devagent-physical = "devagent_physical_engine.cli:main"
 devagent-physical-ai = "devagent_physical_engine.ai_cli:main"
 devagent-commercial = "devagent_physical_engine.commercial_cli:main"
@@ -75,11 +75,14 @@ omit = [
   "src/devagent_physical_engine/ros2/moveit_task_planner.py",
   "src/devagent_physical_engine/ros2/industrial_sequence_planner_v170.py",
   "src/devagent_physical_engine/ros2/industrial_sequence_planner_v171.py",
+  "src/devagent_physical_engine/ros2/industrial_sequence_planner_v172.py",
   "src/devagent_physical_engine/ros2/support_contact_lease_v170.py",
   "src/devagent_physical_engine/ros2/fk_pose.py",
   "src/devagent_physical_engine/ros2/gazebo_pose.py",
   "src/devagent_physical_engine/ros2/workpiece_follow.py",
   "src/devagent_physical_engine/ros2/workpiece_gazebo_follower.py",
+  "src/devagent_physical_engine/ros2/abb_irb1200.py",
+  "src/devagent_physical_engine/ros2/abb_irb1200_preview.py",
 ]
 
 [tool.coverage.report]

{devagent_physical_engine-1.7.1 → devagent_physical_engine-1.8.0}/src/devagent_physical_engine/__init__.py

@@ -217,4 +217,4 @@ install_smooth_motion_hotfix_v162()
 install_industrial_motion_v170()
 install_one_command_runtime()
 
-__version__ = "1.7.1"
+__version__ = "1.8.0"

{devagent_physical_engine-1.7.1 → devagent_physical_engine-1.8.0}/src/devagent_physical_engine/industrial_motion_hotfix_v170.py

@@ -24,23 +24,24 @@ def _patch_constructor(cls) -> None:
 
 
 def install_industrial_motion_v170() -> None:
-    """Route report-test execution through v1.7 industrial motion runtime.
+    """Route report-test execution through the qualified v1.7 industrial runtime.
 
-    v1.7.1 additionally swaps the target-stack planner symbol used by the visual
-    runner for a fail-closed adaptive-blend planner. It retries only generic Pilz
-    sequence planning failures with smaller internal blend radii; no geometry,
-    collision, speed, scene, or execution-authority constraint is relaxed.
+    v1.7.1 retries only generic Pilz sequence-planning failures with smaller internal
+    blend radii. v1.7.2 retains that bounded fallback and, if transfer/place still
+    fails at zero blend, runs plan-only transfer PTP and place LIN probes against the
+    same attached-workpiece scene. The probes only improve diagnosis; they never grant
+    execution authority or relax geometry, collision, speed, scene, or Twin lineage.
     """
 
     from . import report_test_execution as base_exec
     from . import report_test_execution_strict as strict_exec
     from .ros2 import report_test_visualization as base_visual
     from .ros2 import report_test_visualization_v170 as visual_v170
-    from .ros2.industrial_sequence_planner_v171 import (
-        PilzIndustrialSequencePlannerV171,
+    from .ros2.industrial_sequence_planner_v172 import (
+        PilzIndustrialSequencePlannerV172,
     )
 
-    visual_v170.PilzIndustrialSequencePlannerV170 = PilzIndustrialSequencePlannerV171
+    visual_v170.PilzIndustrialSequencePlannerV170 = PilzIndustrialSequencePlannerV172
 
     base_exec.run_report_test_visualization = run_report_test_visualization_v170
     strict_exec.run_report_test_visualization = run_report_test_visualization_v170

{devagent_physical_engine-1.7.1 → devagent_physical_engine-1.8.0}/src/devagent_physical_engine/one_command_v16.py

@@ -13,6 +13,10 @@ from .commercial_repository import CommercialProjectRepository
 from .one_command_v15 import OneCommandVerifier as OneCommandVerifierV15
 from .one_command_v15 import SimpleVerifyResult
 from .report_test_execution_strict import ReportTestFullExecutor
+from .robot_support_v18 import ensure_v18_robot_profiles
+
+
+ensure_v18_robot_profiles()
 
 
 class ProfileGatedReportTestFullExecutor(ReportTestFullExecutor):
@@ -23,6 +27,10 @@ class ProfileGatedReportTestFullExecutor(ReportTestFullExecutor):
     entering UR5e-specific request materialization so that One-Command can continue
     producing bounded campaign/FAT evidence instead of surfacing an executor
     traceback.
+
+    Exact profiles may expose an experimental visual runtime (for example ABB
+    IRB1200) while remaining blocked here until that runtime passes measured FAT
+    qualification. Availability is deliberately not treated as evidence authority.
     """
 
     def execute(

{devagent_physical_engine-1.7.1 → devagent_physical_engine-1.8.0}/src/devagent_physical_engine/qualification.py

@@ -9,6 +9,17 @@ from .robots import CATALOG
 from .simulation import DeterministicSimulationBackend
 
 
+# This is the original v1 cross-vendor *family* corpus. Exact model profiles added
+# later (for example abb_irb1200) must not silently change the historical smoke
+# baseline or turn a family-level qualification into a count of registry entries.
+V1_ROBOT_FAMILIES = (
+    "ur5e",
+    "fanuc_crx",
+    "kuka_kr",
+    "abb_irb",
+)
+
+
 @dataclass(frozen=True, slots=True)
 class QualificationCaseResult:
     robot: str
@@ -35,10 +46,24 @@ def run_cross_vendor_smoke() -> QualificationSummary:
     supervisor = ExecutionSupervisor()
     goal = Goal("G1", "load", "P17", "conveyor_a", "cnc_04")
     results: list[QualificationCaseResult] = []
-    for robot_name, factory in CATALOG.items():
+    for robot_name in V1_ROBOT_FAMILIES:
+        factory = CATALOG[robot_name]
         resource = factory()
         graph = planner.plan(goal, [resource])
         world = WorldState(facts={"robot_ready", "object_available"})
-        report = supervisor.execute(graph, [resource], world, DeterministicSimulationBackend(), ExecutionMode.SIMULATION)
-        results.append(QualificationCaseResult(robot_name, report.completed, len(report.executed_tasks), len(report.issues)))
+        report = supervisor.execute(
+            graph,
+            [resource],
+            world,
+            DeterministicSimulationBackend(),
+            ExecutionMode.SIMULATION,
+        )
+        results.append(
+            QualificationCaseResult(
+                robot_name,
+                report.completed,
+                len(report.executed_tasks),
+                len(report.issues),
+            )
+        )
     return QualificationSummary(tuple(results))

devagent_physical_engine-1.8.0/src/devagent_physical_engine/robot_support_v18.py

@@ -0,0 +1,97 @@
+from __future__ import annotations
+
+from .robot_platform import (
+    ARM_CAPABILITIES,
+    BUILTIN_ROBOT_PROFILES,
+    QualificationState,
+    RobotProfile,
+    RobotQualification,
+    RosIntegration,
+    SimulationTier,
+)
+
+ABB_IRB1200_PROFILE_KEY = "abb_irb1200"
+
+
+def _abb_irb1200_profile() -> RobotProfile:
+    """Exact ABB IRB 1200-5/0.9 profile.
+
+    The visual/physics runtime remains EXPERIMENTAL until target-workstation
+    acceptance proves the exact Gazebo/MoveIt integration. This makes the runtime
+    available without prematurely upgrading FAT evidence authority.
+    """
+
+    return RobotProfile(
+        key=ABB_IRB1200_PROFILE_KEY,
+        vendor="abb",
+        model="irb1200_5_90",
+        capabilities=ARM_CAPABILITIES,
+        qualification=RobotQualification(
+            {
+                SimulationTier.DETERMINISTIC: QualificationState.QUALIFIED,
+                SimulationTier.KINEMATIC: QualificationState.EXPERIMENTAL,
+                SimulationTier.PHYSICS: QualificationState.EXPERIMENTAL,
+                SimulationTier.VISUAL: QualificationState.EXPERIMENTAL,
+            }
+        ),
+        ros=RosIntegration(
+            ros2_compatible=True,
+            joint_state_interface="/joint_states",
+            trajectory_interface="joint_trajectory_controller/follow_joint_trajectory",
+            robot_description="abb_irb1200_support",
+            moveit_config="abb_irb1200_5_90_moveit_config",
+            hardware_driver="abb_ros2",
+        ),
+        aliases=(
+            "abb irb1200",
+            "abb irb 1200",
+            "abb irb 1200-5/0.9",
+            "abb irb1200-5/0.9",
+            "irb1200",
+            "irb1200 5 90",
+        ),
+        legacy_qualification="experimental",
+        notes=(
+            "Exact ABB IRB 1200-5/0.9 ROS2 profile.",
+            "Gazebo Harmonic + gz_ros2_control + MoveIt/RViz runtime is available "
+            "as an experimental visual path until laptop acceptance is recorded.",
+            "Visual preview motion is not FAT measurement evidence and real execution "
+            "remains locked.",
+        ),
+    )
+
+
+def ensure_v18_robot_profiles() -> None:
+    """Idempotently extend the built-in registry with exact v1.8 profiles."""
+
+    try:
+        BUILTIN_ROBOT_PROFILES.get(ABB_IRB1200_PROFILE_KEY)
+    except Exception:
+        BUILTIN_ROBOT_PROFILES.register(_abb_irb1200_profile())
+
+
+def robot_support_level(profile: RobotProfile) -> str:
+    """Return a user-facing support label that never overstates qualification."""
+
+    physics = profile.qualification.state_for(SimulationTier.PHYSICS)
+    visual = profile.qualification.state_for(SimulationTier.VISUAL)
+    kinematic = profile.qualification.state_for(SimulationTier.KINEMATIC)
+
+    if physics is QualificationState.QUALIFIED and visual is QualificationState.QUALIFIED:
+        return "SIMULATION_SUPPORTED"
+    if physics is QualificationState.EXPERIMENTAL and visual is QualificationState.EXPERIMENTAL:
+        return "SIMULATION_AVAILABLE_EXPERIMENTAL"
+    if visual is QualificationState.QUALIFIED or kinematic is QualificationState.QUALIFIED:
+        return "VISUAL_SUPPORTED"
+    if visual is QualificationState.EXPERIMENTAL or kinematic is QualificationState.EXPERIMENTAL:
+        return "VISUAL_AVAILABLE_EXPERIMENTAL"
+    return "MODEL_ONLY"
+
+
+ensure_v18_robot_profiles()
+
+__all__ = [
+    "ABB_IRB1200_PROFILE_KEY",
+    "ensure_v18_robot_profiles",
+    "robot_support_level",
+]

{devagent_physical_engine-1.7.1 → devagent_physical_engine-1.8.0}/src/devagent_physical_engine/robots.py

@@ -9,8 +9,10 @@ from .robot_platform import (
     RobotPlatformError,
     RobotProfileRegistry,
 )
+from .robot_support_v18 import ensure_v18_robot_profiles
 
 
+ensure_v18_robot_profiles()
 PROFILE_REGISTRY: RobotProfileRegistry = BUILTIN_ROBOT_PROFILES
 
 
@@ -34,6 +36,10 @@ def abb_irb(resource_id: str = "robot_01") -> Resource:
     return _resource("abb_irb", resource_id)
 
 
+def abb_irb1200(resource_id: str = "robot_01") -> Resource:
+    return _resource("abb_irb1200", resource_id)
+
+
 class _RegistryBackedCatalog(Mapping[str, Callable[..., Resource]]):
     """Compatibility view for older core code that expects CATALOG[key]().