feat: APS-AgentMesh adapter — structural authorization gates

packages/agentmesh-integrations/aps-agentmesh/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) Microsoft Corporation.
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

packages/agentmesh-integrations/aps-agentmesh/README.md

@@ -0,0 +1,103 @@
+# APS-AgentMesh Integration
+
+AgentMesh adapter for the [Agent Passport System](https://github.com/aeoess/agent-passport-system) (APS). Bridges APS structural authorization into AGT's PolicyEngine as external trust signals.
+
+## Architecture
+
+APS governs **between** processes: cryptographic proof of authorization scope via Ed25519 delegation chains with monotonic narrowing.
+
+AGT governs **inside** the process: policy evaluation, trust scoring, execution rings.
+
+Together: APS structural authorization is a **hard constraint** (gate). AGT behavioral trust scoring is a **soft signal**.
+
+## Components
+
+| Component | Purpose |
+|-----------|---------|
+| `APSPolicyGate` | Injects APS PolicyDecision into AGT evaluation context |
+| `APSTrustBridge` | Maps APS passport grades (0-3) to AGT trust scores (0-1000) |
+| `APSScopeVerifier` | Validates APS delegation scope chains for task assignment |
+| `aps_context()` | Builds AGT-compatible context dict from APS artifacts |
+| `verify_aps_signature()` | Ed25519 signature verification for APS artifacts |
+
+## Passport Grades → Trust Scores
+
+| Grade | Label | Trust Score | Meaning |
+|-------|-------|-------------|---------|
+| 0 | self_signed | 100 | Bare Ed25519 keypair |
+| 1 | issuer_countersigned | 400 | AEOESS processed the request |
+| 2 | runtime_bound | 700 | Challenge-response + infrastructure attestation |
+| 3 | principal_bound | 900 | Runtime + verified human/org principal |
+
+## Usage
+
+### As AGT PolicyEngine context
+
+```python
+from aps_agentmesh import APSPolicyGate
+
+gate = APSPolicyGate()
+
+# APS PolicyDecision (from APS gateway or MCP server)
+aps_decision = {
+    "verdict": "permit",
+    "scopeUsed": "deploy.staging",
+    "agentId": "claude-operator",
+    "delegationId": "del-abc123",
+}
+
+# Build AGT-compatible context
+context = gate.build_context(aps_decision, passport_grade=2)
+
+# Pass to AGT PolicyEngine
+decision = policy_engine.evaluate("deploy.staging", context)
+```
+
+### AGT policy rule consuming APS
+
+```yaml
+- name: require-aps-authorization
+  type: capability
+  conditions:
+    aps_decision.verdict: "permit"
+  allowed_actions:
+    - "deploy.*"
+```
+
+### Trust bridging
+
+```python
+from aps_agentmesh import APSTrustBridge
+
+bridge = APSTrustBridge()
+
+# Grade 2 (runtime-bound) → 700 trust score
+score = bridge.grade_to_score(passport_grade=2)
+
+# Check minimum threshold
+if bridge.meets_threshold(passport_grade=1, min_score=500):
+    print("Insufficient attestation for this action")
+```
+
+### Scope verification
+
+```python
+from aps_agentmesh import APSScopeVerifier
+
+verifier = APSScopeVerifier()
+ok, reason = verifier.verify(
+    scope_chain=delegation_json,
+    required_scope="commerce:checkout",
+    required_spend=49.99,
+)
+if not ok:
+    print(f"Denied: {reason}")
+```
+
+## APS Resources
+
+- SDK: [npm](https://www.npmjs.com/package/agent-passport-system) (v1.29.1, 1,919 tests)
+- Python SDK: [PyPI](https://pypi.org/project/agent-passport-system/) (v0.7.0)
+- MCP Server: [npm](https://www.npmjs.com/package/agent-passport-system-mcp) (v2.19.0, 125 tools)
+- Docs: [aeoess.com](https://aeoess.com)
+- Paper: [Zenodo DOI 10.5281/zenodo.19260073](https://doi.org/10.5281/zenodo.19260073)

packages/agentmesh-integrations/aps-agentmesh/SECURITY.md

@@ -0,0 +1,3 @@
+# Security
+
+For security issues, please refer to the [main SECURITY.md](../../SECURITY.md).

packages/agentmesh-integrations/aps-agentmesh/aps_agentmesh/__init__.py

@@ -0,0 +1,36 @@
+"""
+APS-AgentMesh Integration — Structural authorization for AgentMesh agents.
+
+The Agent Passport System (APS) provides cryptographic identity, scoped delegation
+chains with monotonic narrowing, and signed policy decisions. This adapter makes
+APS artifacts consumable by AGT's PolicyEngine as external trust signals.
+
+Architecture:
+  APS governs BETWEEN processes (cryptographic proof of authorization scope).
+  AGT governs INSIDE the process (policy evaluation, trust scoring, execution rings).
+  Together: APS structural authorization is a hard gate, AGT behavioral trust is a soft signal.
+
+Provides:
+  - APSPolicyGate: Injects APS PolicyDecision into AGT evaluation context
+  - APSTrustBridge: Maps APS passport grades (0-3) to AGT trust scores (0-1000)
+  - APSScopeVerifier: Validates APS delegation scope chains
+  - verify_aps_signature: Ed25519 signature verification for APS artifacts
+"""
+
+from .adapter import (
+    APSPolicyGate,
+    APSTrustBridge,
+    APSScopeVerifier,
+    aps_context,
+    verify_aps_signature,
+    GRADE_TO_TRUST_SCORE,
+)
+
+__all__ = [
+    "APSPolicyGate",
+    "APSTrustBridge",
+    "APSScopeVerifier",
+    "aps_context",
+    "verify_aps_signature",
+    "GRADE_TO_TRUST_SCORE",
+]