agindow 0.1.0a0__tar.gz

agindow-0.1.0a0/.gitignore

@@ -0,0 +1,5 @@
+.venv/
+dist/
+.mypy_cache/
+__pycache__/
+src/agindow/_guest.wasm

agindow-0.1.0a0/LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2026 AGINDOW
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

agindow-0.1.0a0/NOTICE

@@ -0,0 +1,36 @@
+AGINDOW
+Copyright 2026 AGINDOW
+
+This product includes software developed at AGINDOW.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not
+use these files except in compliance with the License. A copy of the License
+is distributed with this work, and is also available at:
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+--------------------------------------------------------------------------------
+Trademark notice
+--------------------------------------------------------------------------------
+
+"AGINDOW" is a trademark of AGINDOW. The Apache License, Version 2.0 does
+**not** grant permission to use this name, the AGINDOW word mark, or any other
+trademark or logo of the Licensor. See Section 6 of the License.
+
+The License permits use of the name only as reasonably necessary to describe
+the origin of the work and to reproduce the content of this NOTICE file.
+
+Vendors and redistributors who wish to refer to the project by name in their
+own documentation or user interfaces should write something like "built with
+AGINDOW" rather than implying endorsement or affiliation. Please contact the
+Licensor for permission to use the mark in any other way.
+
+--------------------------------------------------------------------------------
+Scope
+--------------------------------------------------------------------------------
+
+This NOTICE applies to the Work distributed under the Apache License, Version
+2.0. Portions of this repository designated as proprietary in
+`tasks/TECH_STACK.md` §6 — the control plane, registry, and dashboard — are not
+covered by the Apache License and are licensed separately. Those portions do
+not exist in this repository yet.

agindow-0.1.0a0/PKG-INFO

@@ -0,0 +1,133 @@
+Metadata-Version: 2.5
+Name: agindow
+Version: 0.1.0a0
+Summary: API drift firewall for AI agents on Python (Rust/WASM)
+Project-URL: Homepage, https://github.com/agindow/sdk
+Project-URL: Repository, https://github.com/agindow/sdk
+Project-URL: Issues, https://github.com/agindow/sdk/issues
+License-Expression: Apache-2.0
+License-File: LICENSE
+License-File: NOTICE
+Keywords: agents,ai,api,drift,firewall,guard,llm,observability,sdk,wasm
+Requires-Python: >=3.11
+Requires-Dist: wasmtime==46.0.1
+Description-Content-Type: text/markdown
+
+# agindow (Python)
+
+API drift firewall for AI agents on Python 3.11+ — repair drift in-flight, prove
+the write was correct. **Pre-alpha** on PyPI as `agindow`. Loads the **identical
+WASM bytes** built for `agindow` through Wasmtime. Python provides
+callable/configuration/availability glue — not a second drift engine.
+
+The default engine detects nothing until you configure protection. Statistical
+windows require explicit local baseline/live selections.
+
+## Install
+
+```bash
+pip install agindow
+# or: uv add agindow
+```
+
+The wheel includes the prebuilt WASM guest. You do **not** need Rust, Node, or a
+monorepo checkout to use the SDK.
+
+## Decorators, async and contexts
+
+```python
+from agindow import Agindow, guard
+
+@guard(mode="repair", tool="customers.create", operation="write")
+def create_customer(email: str) -> dict[str, str]:
+    return {"email": email}
+
+sdk = Agindow(mode="observe")
+
+@sdk.guard(tool="customers.read", operation="read")
+async def read_customer() -> dict[str, str]:
+    return {"id": "fixture"}
+
+with sdk.context(mode="strict"):
+    # Only calls through this SDK's guards use the scoped configuration.
+    pass
+# `async with sdk.context(...)` is supported as well.
+sdk.close()
+```
+
+Context managers scope guarded calls; they cannot intercept arbitrary Python
+statements. Overrides use ContextVar, nest and remain isolated across async
+tasks. Each call freezes configuration until completion. `configure(mode="repair")`
+changes future calls. `sdk.ready` is a synchronous boolean after local loading
+(unlike Node’s startup Promise); `sdk.load_error` retains a loading failure.
+Loading failures and retired guests preserve upstream results and emit one
+bypass per call. `close()` is safe for shutdown; subsequent calls bypass.
+
+The wrapper preserves `__name__`, documentation, annotations, signatures,
+bound receivers and arguments. ParamSpec retains Pydantic model types without
+a mandatory Pydantic dependency. Model results retain their original identity;
+only a detached `model_dump(mode="json")` projection enters inspection.
+
+## Configuration and feedback
+
+Supported fields (snake_case): `mode`, `fail_closed`, `api_key`, `endpoint`,
+`allowed_tools` (`"*"` or a tuple), `deny_tools`, `buffer` and `on_event`.
+Resolution: explicit keywords > environment > nearest JSON `.agindowrc` >
+defaults. Shared env vars match Node (`AGINDOW_MODE`, `AGINDOW_KEY`,
+`AGINDOW_ENDPOINT`, `AGINDOW_BUFFER`, `AGINDOW_FAIL_CLOSED`). Invalid explicit
+or environment values raise `AgindowSetupError` before a call.
+
+```python
+from agindow import AgindowDriftError
+
+try:
+    create_customer("fixture@example.test")
+except AgindowDriftError as error:
+    print(error.code, error.tool, error.drift_class, error.drift_id)
+    print(error.hint, error.audit_url)
+```
+
+Request halts dispatch zero times. Response halts retain `error.outcome` and
+warn against blind replay. Upstream exceptions survive as `error.__cause__`.
+Missing audit IDs/URLs remain empty rather than fabricated.
+
+`on(type, handler)` returns an unsubscribe callback; `recent()` returns an
+immutable tuple of retained events. `buffer=0` disables retention. This is local
+telemetry, not hosted audit delivery.
+
+## Shared-core validation and limits
+
+Parity fixtures run both public wrappers against the same WASM guest. The
+paired local agent-loop benchmark gates added p50 <100 µs on the default
+NullEngine only (excludes model/network/import/statistical work).
+
+Python `wrap` currently accepts callables. Node-specific tool/client/HTTP
+adapters, Nango codegen, OTel spans, report CLI and remote configuration/audit
+transports are outside this package. No Python framework adapter or streaming
+qualification is claimed. Inspection is bounded to detached JSON projections.
+
+## Remote fail-open API
+
+Local decorators stay offline. Remote execution is a separate HTTP API using
+stdlib HTTP transport; it never redirects or retries through AGINDOW:
+
+```python
+from agindow import RemoteEnvelope, RemoteOptions, guard_remote
+
+request = RemoteEnvelope("vendor.read", "example", "read", {}, "observe")
+result = guard_remote(
+    lambda envelope: {"fallback": True},
+    request,
+    RemoteOptions("http://127.0.0.1:8789/v1/guard"),
+)
+```
+
+`RemoteResponse` retains the original status/headers and streaming `body`.
+Any 5xx/refusal/timeout/reset invokes the original callable once and emits one
+bypass. Writes/unknown operations require the same upstream idempotency key in
+the envelope and fallback. `guard_remote_async` runs the blocking API in a
+worker thread for a synchronous fallback callable.
+
+Opt-in machine-identity helpers stay credential-free in embedded mode; they do
+not enable a hosted identity service by themselves. Sibling Node package:
+`agindow`.

agindow-0.1.0a0/README.md

@@ -0,0 +1,118 @@
+# agindow (Python)
+
+API drift firewall for AI agents on Python 3.11+ — repair drift in-flight, prove
+the write was correct. **Pre-alpha** on PyPI as `agindow`. Loads the **identical
+WASM bytes** built for `agindow` through Wasmtime. Python provides
+callable/configuration/availability glue — not a second drift engine.
+
+The default engine detects nothing until you configure protection. Statistical
+windows require explicit local baseline/live selections.
+
+## Install
+
+```bash
+pip install agindow
+# or: uv add agindow
+```
+
+The wheel includes the prebuilt WASM guest. You do **not** need Rust, Node, or a
+monorepo checkout to use the SDK.
+
+## Decorators, async and contexts
+
+```python
+from agindow import Agindow, guard
+
+@guard(mode="repair", tool="customers.create", operation="write")
+def create_customer(email: str) -> dict[str, str]:
+    return {"email": email}
+
+sdk = Agindow(mode="observe")
+
+@sdk.guard(tool="customers.read", operation="read")
+async def read_customer() -> dict[str, str]:
+    return {"id": "fixture"}
+
+with sdk.context(mode="strict"):
+    # Only calls through this SDK's guards use the scoped configuration.
+    pass
+# `async with sdk.context(...)` is supported as well.
+sdk.close()
+```
+
+Context managers scope guarded calls; they cannot intercept arbitrary Python
+statements. Overrides use ContextVar, nest and remain isolated across async
+tasks. Each call freezes configuration until completion. `configure(mode="repair")`
+changes future calls. `sdk.ready` is a synchronous boolean after local loading
+(unlike Node’s startup Promise); `sdk.load_error` retains a loading failure.
+Loading failures and retired guests preserve upstream results and emit one
+bypass per call. `close()` is safe for shutdown; subsequent calls bypass.
+
+The wrapper preserves `__name__`, documentation, annotations, signatures,
+bound receivers and arguments. ParamSpec retains Pydantic model types without
+a mandatory Pydantic dependency. Model results retain their original identity;
+only a detached `model_dump(mode="json")` projection enters inspection.
+
+## Configuration and feedback
+
+Supported fields (snake_case): `mode`, `fail_closed`, `api_key`, `endpoint`,
+`allowed_tools` (`"*"` or a tuple), `deny_tools`, `buffer` and `on_event`.
+Resolution: explicit keywords > environment > nearest JSON `.agindowrc` >
+defaults. Shared env vars match Node (`AGINDOW_MODE`, `AGINDOW_KEY`,
+`AGINDOW_ENDPOINT`, `AGINDOW_BUFFER`, `AGINDOW_FAIL_CLOSED`). Invalid explicit
+or environment values raise `AgindowSetupError` before a call.
+
+```python
+from agindow import AgindowDriftError
+
+try:
+    create_customer("fixture@example.test")
+except AgindowDriftError as error:
+    print(error.code, error.tool, error.drift_class, error.drift_id)
+    print(error.hint, error.audit_url)
+```
+
+Request halts dispatch zero times. Response halts retain `error.outcome` and
+warn against blind replay. Upstream exceptions survive as `error.__cause__`.
+Missing audit IDs/URLs remain empty rather than fabricated.
+
+`on(type, handler)` returns an unsubscribe callback; `recent()` returns an
+immutable tuple of retained events. `buffer=0` disables retention. This is local
+telemetry, not hosted audit delivery.
+
+## Shared-core validation and limits
+
+Parity fixtures run both public wrappers against the same WASM guest. The
+paired local agent-loop benchmark gates added p50 <100 µs on the default
+NullEngine only (excludes model/network/import/statistical work).
+
+Python `wrap` currently accepts callables. Node-specific tool/client/HTTP
+adapters, Nango codegen, OTel spans, report CLI and remote configuration/audit
+transports are outside this package. No Python framework adapter or streaming
+qualification is claimed. Inspection is bounded to detached JSON projections.
+
+## Remote fail-open API
+
+Local decorators stay offline. Remote execution is a separate HTTP API using
+stdlib HTTP transport; it never redirects or retries through AGINDOW:
+
+```python
+from agindow import RemoteEnvelope, RemoteOptions, guard_remote
+
+request = RemoteEnvelope("vendor.read", "example", "read", {}, "observe")
+result = guard_remote(
+    lambda envelope: {"fallback": True},
+    request,
+    RemoteOptions("http://127.0.0.1:8789/v1/guard"),
+)
+```
+
+`RemoteResponse` retains the original status/headers and streaming `body`.
+Any 5xx/refusal/timeout/reset invokes the original callable once and emits one
+bypass. Writes/unknown operations require the same upstream idempotency key in
+the envelope and fallback. `guard_remote_async` runs the blocking API in a
+worker thread for a synchronous fallback callable.
+
+Opt-in machine-identity helpers stay credential-free in embedded mode; they do
+not enable a hosted identity service by themselves. Sibling Node package:
+`agindow`.

agindow-0.1.0a0/examples/offline.py

@@ -0,0 +1,34 @@
+"""Fictional local evidence: no credentials, vendor calls or production claims."""
+from __future__ import annotations
+
+import json
+from agindow import Agindow, AgindowDriftError
+
+selection: dict[str, object] = {
+    "tenant": "fixture", "connection": "fixture", "vendor": "fixture", "operation": "op", "segment": "fixture",
+    "fields": [{"side": "request", "path": "/value", "baseline": ['{"value":1}'] * 128,
+                "live": ['{"value":100}'] * 128, "enumeration": True}],
+}
+sdk = Agindow(mode="repair", selection=selection)
+if not sdk.ready:
+    raise RuntimeError(f"shared guest unavailable: {sdk.load_error}")
+calls = 0
+
+
+@sdk.guard(tool="op", operation="write")
+def upstream(value: dict[str, int]) -> dict[str, int]:
+    global calls
+    calls += 1
+    return value
+
+
+original = {"value": 100}
+try:
+    upstream(original)
+    raise AssertionError("fixture should halt before dispatch")
+except AgindowDriftError as error:
+    print(json.dumps({"code": error.code, "phase": error.phase, "dispatches": calls}))
+with sdk.context(mode="observe"):
+    assert upstream(original) is original
+print(json.dumps({"observe_preserved_reference": True, "total_dispatches": calls}))
+sdk.close()

agindow-0.1.0a0/pyproject.toml

@@ -0,0 +1,55 @@
+[project]
+name = "agindow"
+version = "0.1.0a0"
+description = "API drift firewall for AI agents on Python (Rust/WASM)"
+readme = "README.md"
+requires-python = ">=3.11"
+license = "Apache-2.0"
+keywords = [
+    "ai",
+    "agents",
+    "api",
+    "drift",
+    "firewall",
+    "llm",
+    "wasm",
+    "guard",
+    "sdk",
+    "observability",
+]
+dependencies = [
+    "wasmtime==46.0.1",
+]
+
+[project.urls]
+Homepage = "https://github.com/agindow/sdk"
+Repository = "https://github.com/agindow/sdk"
+Issues = "https://github.com/agindow/sdk/issues"
+
+[build-system]
+requires = ["hatchling==1.32.4"]
+build-backend = "hatchling.build"
+
+[dependency-groups]
+dev = [
+    "hatchling>=1.27,<2",
+    "mypy>=1.15,<2",
+    "pydantic>=2.11,<3",
+]
+
+[tool.hatch.build.targets.wheel]
+packages = ["src/agindow"]
+
+[tool.hatch.build.targets.wheel.force-include]
+"src/agindow/_guest.wasm" = "agindow/_guest.wasm"
+
+[tool.hatch.build.targets.sdist]
+include = ["src", "tests", "scripts", "examples", "pyproject.toml", "uv.lock", "README.md", "LICENSE", "NOTICE"]
+
+[tool.hatch.build.targets.sdist.force-include]
+"src/agindow/_guest.wasm" = "src/agindow/_guest.wasm"
+
+[tool.mypy]
+python_version = "3.11"
+strict = true
+files = ["src", "tests", "scripts", "examples"]

agindow-0.1.0a0/scripts/bench.py

@@ -0,0 +1,72 @@
+"""Paired local default-engine agent loop; no model/vendor/network latency."""
+from __future__ import annotations
+
+import json
+import math
+import platform
+import sys
+from importlib.metadata import version
+from pathlib import Path
+from statistics import mean, pstdev
+from time import perf_counter_ns
+from typing import Callable
+from agindow import Agindow
+
+
+def quantile(values: list[float], probability: float) -> float:
+    return sorted(values)[math.ceil(len(values) * probability) - 1]
+
+
+def measure() -> dict[str, object]:
+    started = perf_counter_ns()
+    sdk = Agindow()
+    load_us = (perf_counter_ns() - started) / 1000
+    if not sdk.ready:
+        raise RuntimeError(f"guest failed to load: {sdk.load_error}")
+    payload = {"result": "x" * 849}
+    tools = {"lookup": lambda: payload, "verify": lambda: payload}
+    guarded = {name: sdk.guard(fn) for name, fn in tools.items()}
+    steps = 0
+    observation = payload
+
+    def step(target: dict[str, Callable[[], dict[str, str]]]) -> None:
+        nonlocal observation, steps
+        action = "lookup" if (len(observation["result"]) + steps) % 2 == 0 else "verify"
+        observation = target[action]()
+        if observation is not payload:
+            raise RuntimeError("outcome identity lost")
+        steps += 1
+
+    for _ in range(1000):
+        step(tools)
+        step(guarded)
+    raw: dict[str, list[float]] = {"guarded": [], "unguarded": [], "added": []}
+    for index in range(5000):
+        sample: dict[str, float] = {}
+        for kind in (("guarded", "unguarded") if index % 2 == 0 else ("unguarded", "guarded")):
+            started = perf_counter_ns()
+            step(guarded if kind == "guarded" else tools)
+            sample[kind] = (perf_counter_ns() - started) / 1000
+            raw[kind].append(sample[kind])
+        raw["added"].append(sample["guarded"] - sample["unguarded"])
+    summary = {key: {"p50": quantile(values, .5), "p99": quantile(values, .99),
+                     "mean": mean(values), "stddev": pstdev(values)} for key, values in raw.items()}
+    bypasses = sum(event.type == "bypass" for event in sdk.recent())
+    report: dict[str, object] = {"schema": "agindow.python-loop/1", "samples": 5000, "warmups": 1000,
+        "fixture": "local agent tool-selection/observation loop; NullEngine through the shared release WASM",
+        "estimator": "p50 of alternating paired guarded-minus-unguarded steps; nearest-rank quantiles",
+        "steps": steps, "payloadBytes": 862, "bypasses": bypasses, "loadUs": load_us, "summary": summary, "raw": raw,
+        "environment": {"python": sys.version, "wasmtime": version("wasmtime"), "platform": platform.platform(),
+                        "arch": platform.machine(), "cpu": platform.processor()},
+        "exclusions": ["module import before load timer", "model inference", "vendor/network", "statistical paths (TI-53)"]}
+    output = Path(__file__).resolve().parents[4] / "target/python-loop/report.json"
+    output.parent.mkdir(parents=True, exist_ok=True)
+    output.write_text(json.dumps(report, indent=2) + "\n")
+    print(json.dumps({"addedUs": summary["added"], "loadUs": load_us, "report": str(output)}))
+    if summary["added"]["p50"] >= 100 or bypasses:
+        raise RuntimeError("Python default-engine loop budget failed")
+    return report
+
+
+if __name__ == "__main__":
+    measure()