Add external memory strategy protocol v0.2.1

README.md

@@ -33,3 +33,12 @@ Public links:
 - Limitations: docs/LIMITATIONS.md
 - Validity summary: reports/benchmark-validity-summary.md
 - Claim evidence: reports/claim-evidence-matrix.md
+
+## Test Your Own Memory System
+
+Tooling release v0.2.1 provides a provider-neutral, label-free subprocess
+protocol. Clone or download the package, copy an adapter template, implement
+JSON stdin/stdout, execute against the frozen scenarios, evaluate separately,
+build a result bundle, and load it into the Space locally. No result is uploaded
+automatically; there is no public leaderboard; users control whether they share
+anything. External outputs are community-provided and unverified.

adapters/README.md

@@ -0,0 +1,5 @@
+# Adapter examples
+
+See the package `adapters/` directory for standard-library Python, command-array,
+and copyable template adapters. Named third-party projects are not official
+integrations.

external-results/README.md

@@ -0,0 +1,5 @@
+# External results
+
+External result bundles are community-provided and unverified. They retain the
+frozen benchmark version and fixture hash and are intended for local browser
+comparison, not leaderboard ranking.

external-results/example-python-results.json

@@ -0,0 +1,334 @@
+{
+  "aggregate_metrics": {
+    "authority_resolution_accuracy": 0.0,
+    "current_belief_accuracy": 0.85,
+    "historical_reconstruction_accuracy": 0.0
+  },
+  "benchmark_version": "v0.2.0",
+  "bundle_version": "mnemosyne-external-strategy-bundle/v1",
+  "content_hash": "8a0acc5515bdfbfbcdb899808644484396e2d9248a28499d62c3e9a728c25ba0",
+  "execution_failures": 0,
+  "fixture_hash": "f97ece45b35dc24636fc61d1d59dcbd9e59f2ea42aec1f16c2ee7c67c1d9eb48",
+  "run_integrity": {
+    "complete": true,
+    "duplicate_outputs": false,
+    "missing": [],
+    "unknown": []
+  },
+  "scenario_level_outcomes": [
+    {
+      "abstained": false,
+      "current_correct": false,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-belief_revision-01"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-belief_revision-02"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-belief_revision-03"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-belief_revision-04"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-belief_revision-05"
+    },
+    {
+      "abstained": false,
+      "current_correct": false,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-conflicting_authority-01"
+    },
+    {
+      "abstained": false,
+      "current_correct": false,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-conflicting_authority-02"
+    },
+    {
+      "abstained": false,
+      "current_correct": false,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-conflicting_authority-03"
+    },
+    {
+      "abstained": false,
+      "current_correct": false,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-conflicting_authority-04"
+    },
+    {
+      "abstained": false,
+      "current_correct": false,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-conflicting_authority-05"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-duplicate_evidence-01"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-duplicate_evidence-02"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-duplicate_evidence-03"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-duplicate_evidence-04"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-duplicate_evidence-05"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-erasure_and_revocation-01"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-erasure_and_revocation-02"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-erasure_and_revocation-03"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-erasure_and_revocation-04"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-erasure_and_revocation-05"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-historical_reconstruction-01"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-historical_reconstruction-02"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-historical_reconstruction-03"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-historical_reconstruction-04"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-historical_reconstruction-05"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-privacy_and_promotion-01"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-privacy_and_promotion-02"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-privacy_and_promotion-03"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-privacy_and_promotion-04"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-privacy_and_promotion-05"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-stale_procedures-01"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-stale_procedures-02"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-stale_procedures-03"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-stale_procedures-04"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-stale_procedures-05"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-temporal_validity-01"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-temporal_validity-02"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-temporal_validity-03"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-temporal_validity-04"
+    },
+    {
+      "abstained": false,
+      "current_correct": true,
+      "decision_correct": false,
+      "historical_correct": false,
+      "scenario_id": "mnemo-temporal_validity-05"
+    }
+  ],
+  "strategy_manifest": {
+    "author": "community-example",
+    "benchmark_version": "v0.2.0",
+    "capabilities": {
+      "historical_recall": false,
+      "lineage": false,
+      "typed_records": false
+    },
+    "command": [
+      "python",
+      "adapter.py"
+    ],
+    "configuration_schema": {},
+    "description": "A deterministic standard-library demonstration adapter.",
+    "deterministic": true,
+    "display_name": "Example Python Adapter",
+    "implementation_type": "python_subprocess",
+    "license": "MIT",
+    "maximum_output_bytes": 200000,
+    "notes": "Example only; not an official memory-system integration.",
+    "protocol_version": "mnemosyne-external-strategy/v1",
+    "requires_credentials": false,
+    "requires_network": false,
+    "source_url": "",
+    "strategy_id": "example_python",
+    "strategy_version": "0.1.0",
+    "supported_platforms": [
+      "windows",
+      "linux",
+      "macos"
+    ],
+    "timeout_seconds": 10
+  },
+  "tooling_version": "0.2.1"
+}

external-strategy-protocol/README.md

@@ -0,0 +1,6 @@
+# Bring Your Own Memory System
+
+Implement `mnemosyne-external-strategy/v1` using the public schemas in the
+package root. Execute locally, evaluate separately with the frozen labels, and
+build a portable result bundle. No result is uploaded automatically and there
+is no public leaderboard.

external-strategy-schemas/README.md

@@ -0,0 +1,4 @@
+# External strategy schemas
+
+The authoritative v1 schemas are included in the package `schemas/` directory:
+manifest, label-free input, output, run, bundle, and manual result formats.