@vorionsys/atsf-core 0.2.2 → 0.2.3

package/CHANGELOG.md

@@ -62,6 +62,6 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Fastify-based governance API server (`createServer`, `startServer`)
 - Comprehensive test suite (401+ tests)
 
-[0.2.1]: https://github.com/voriongit/vorion/compare/atsf-core-v0.2.0...atsf-core-v0.2.1
-[0.2.0]: https://github.com/voriongit/vorion/compare/atsf-core-v0.1.0...atsf-core-v0.2.0
-[0.1.0]: https://github.com/voriongit/vorion/releases/tag/atsf-core-v0.1.0
+[0.2.1]: https://github.com/vorionsys/vorion/compare/atsf-core-v0.2.0...atsf-core-v0.2.1
+[0.2.0]: https://github.com/vorionsys/vorion/compare/atsf-core-v0.1.0...atsf-core-v0.2.0
+[0.1.0]: https://github.com/vorionsys/vorion/releases/tag/atsf-core-v0.1.0

package/README.md

@@ -1,6 +1,6 @@
 # @vorionsys/atsf-core
 
-Agentic Trust Scoring Framework (ATSF) -- the core runtime for AI agent governance, trust scoring, and policy enforcement. Implements the complete 8-tier trust model (T0-T7) on a 0-1000 scale with behavioral signal processing, time-based decay, recovery mechanics, and immutable audit trails.
+Agentic Trust Scoring Framework (ATSF) -- the core runtime for AI agent governance, trust scoring, and policy enforcement. Implements the complete 8-tier trust model (T0-T7) on a 0-1000 scale with behavioral signal processing, time-based Readiness Degree adjustments, recovery mechanics, and immutable audit trails.
 
 ## Installation
 
@@ -19,8 +19,8 @@ ATSF continuously evaluates agent behavior across multiple dimensions and assign
 Key principles:
 
 - **Trust is earned, not granted.** Agents start at low trust and must demonstrate competence to advance.
-- **Trust decays over time.** Idle agents lose trust; active, well-behaved agents maintain or gain it.
-- **Trust loss is asymmetric.** Trust is hard to gain and easy to lose (10:1 ratio per ATSF v2.0).
+- **Readiness adjusts over time.** Idle agents can lose readiness; active, well-behaved agents maintain or gain it.
+- **Readiness reduction is asymmetric.** Trust dynamics are configurable by policy profile, with stricter readiness and recovery behavior at higher assurance requirements.
 - **Every decision is auditable.** An immutable proof chain records all governance decisions.
 
 ## The 8-Tier Trust Model (T0-T7)
@@ -78,7 +78,7 @@ await engine.recordSignal({
   metadata: { task: 'data-analysis' },
 });
 
-// Get the updated score (includes automatic decay calculation)
+// Get the updated score (includes automatic Readiness Degree adjustment)
 const record = await engine.getScore('agent-001');
 console.log(record?.score, record?.level);
 ```
@@ -94,7 +94,8 @@ const engine = createTrustEngine({
   decayRate: 0.01,                 // 1% decay per interval
   decayIntervalMs: 60000,         // 1-minute intervals
   failureThreshold: 0.3,          // Signals below 0.3 = failure
-  acceleratedDecayMultiplier: 3.0, // 3x decay on repeated failures
+  acceleratedDecayMultiplier: 1.0, // No extra decay multiplier by default
+  readinessMode: 'checkpoint_schedule', // Preferred neutral mode naming
   successThreshold: 0.7,          // Signals above 0.7 = success
   recoveryRate: 0.02,             // 2% recovery per success signal
 });
@@ -166,7 +167,7 @@ engine.on('trust:tier_changed', (event) => {
 engine.on('trust:failure_detected', (event) => {
   console.log(`Failure #${event.failureCount} for ${event.entityId}`);
   if (event.acceleratedDecayActive) {
-    console.log('Accelerated decay is now active (3x normal rate)');
+    console.log('Additional decay multiplier is active for this policy profile');
   }
 });
 
@@ -370,17 +371,82 @@ Import from the package root or via deep imports:
 | `trust:signal_recorded` | Behavioral signal recorded |
 | `trust:score_changed` | Score changes by 5+ points |
 | `trust:tier_changed` | Entity promoted or demoted |
-| `trust:decay_applied` | Trust decayed (includes `accelerated` flag) |
+| `trust:readiness_adjusted` | Readiness Degree adjusted (primary neutral event) |
+| `trust:freshness_adjusted` | Backward-compatible readiness alias |
+| `trust:decay_applied` | Legacy compatibility alias (includes `accelerated` flag) |
 | `trust:failure_detected` | Signal value below failure threshold |
 | `trust:recovery_applied` | Recovery from successful signal |
 | `trust:recovery_milestone` | Tier restored, full recovery, or accelerated recovery earned |
 | `trust:*` | Wildcard -- all trust events |
 
+## Wave 1 API Stability
+
+The following modules are **stable for Wave 1** on the `0.2.x` series and covered by the semver guarantee. Anything not listed here is internal and subject to change without notice.
+
+### Stable (≥ v0.2.2)
+
+| Module | Key Exports |
+|--------|-------------|
+| `trust-engine` | `TrustEngine`, `createTrustEngine`, `TRUST_THRESHOLDS`, `TRUST_LEVEL_NAMES`, `SIGNAL_WEIGHTS` |
+| `basis` | `createEvaluator`, `RuleEvaluator`, `parseNamespace`, `validateRule`, `validateNamespace` |
+| `intent` | `createIntentService`, `IntentService` |
+| `enforce` | `createEnforcementService`, `EnforcementService` |
+| `proof` | `createProofService`, `ProofService` |
+| `governance` | `createGovernanceEngine`, `GovernanceEngine`, `createGovernanceRule` |
+| `persistence` | `createFileProvider`, `createMemoryProvider` |
+| `phase6` | `createPhase6TrustEngine`, `BASIS_CANONICAL_PRESETS`, `TrustTier`, `AgentRole`, `ContextType` |
+| `layers` | `createSecurityPipeline`, `BaseSecurityLayer` |
+| `langchain` | `createTrustAwareExecutor`, `createTrustTools` |
+| `sandbox-training` | `PromotionService`, `CHALLENGE_CATALOG` |
+| `chain` | `createChainAnchor`, `computeMerkleRoot`, `computeMerkleProof`, `sha256` |
+
+### Beta (functional — API shape may change before Wave 2)
+
+| Module | Notes |
+|--------|-------|
+| `intent-gateway` | `IntentGateway`, `createIntentGateway` — jurisdiction selectors not yet final |
+| `crewai` | CrewAI integration utilities |
+| `arbitration` | `MultiAgentTrustArbitrator` — multi-agent quorum logic stabilizing |
+| `containment` | `ProgressiveContainmentService` — protocol steps may be extended |
+
+---
+
+## Deprecated Exports
+
+The following are present for backwards compatibility only. Migrate before v1.0.0.
+
+### `ACI_CANONICAL_PRESETS` → `BASIS_CANONICAL_PRESETS`
+
+`ACI_CANONICAL_PRESETS` is a legacy alias created during the ACI → BASIS nomenclature rename. It points to exactly the same object as `BASIS_CANONICAL_PRESETS` — there is no behavioral difference.
+
+```typescript
+// ❌ Deprecated — do not use in new code
+import { phase6 } from '@vorionsys/atsf-core';
+const preset = phase6.ACI_CANONICAL_PRESETS['basis:preset:balanced'];
+
+// ✅ Correct Wave 1 pattern
+import { phase6 } from '@vorionsys/atsf-core';
+const preset = phase6.BASIS_CANONICAL_PRESETS['basis:preset:balanced'];
+```
+
+| Milestone | Action |
+|-----------|--------|
+| `v0.3.0` | `@deprecated` JSDoc annotation added — TypeScript will surface a warning on all call sites |
+| `v1.0.0` | Export removed (Wave 2 release) |
+
+**Migration:** search-and-replace only — no behavioral change.
+
+### `ACIClient`, `ACIError`, `createACIClient`, `createLocalACIClient` (in `@vorionsys/car-client`)
+
+Same ACI → CAR rename — aliases in `@vorionsys/car-client` follow the identical removal timeline.
+
+---
+
 ## Architecture
 
 ```
 @vorionsys/atsf-core
-  |-- trust-engine/     Trust scoring with 8-tier model, decay, and recovery
+  |-- trust-engine/     Trust scoring with 8-tier model, readiness adjustments, and recovery
   |-- basis/            BASIS rule evaluation engine
   |-- intent/           Intent submission and lifecycle tracking
   |-- enforce/          Policy decision point (allow/deny/escalate)
@@ -409,7 +475,7 @@ Import from the package root or via deep imports:
 
 ## Testing
 
-The package has comprehensive test coverage with **401+ tests** covering trust scoring, decay mechanics, recovery paths, governance pipelines, security layers, and edge cases.
+The package has comprehensive test coverage with **401+ tests** covering trust scoring, readiness-adjustment mechanics, recovery paths, governance pipelines, security layers, and edge cases.
 
 ```bash
 # Run all tests
@@ -427,6 +493,6 @@ Copyright 2024-2026 Vorion
 
 ## Links
 
-- **Repository:** [github.com/voriongit/vorion](https://github.com/voriongit/vorion/tree/main/packages/atsf-core)
+- **Repository:** [github.com/vorionsys/vorion](https://github.com/vorionsys/vorion/tree/main/packages/atsf-core)
 - **Homepage:** [vorion.org](https://vorion.org)
-- **Issues:** [github.com/voriongit/vorion/issues](https://github.com/voriongit/vorion/issues)
+- **Issues:** [github.com/vorionsys/vorion/issues](https://github.com/vorionsys/vorion/issues)

package/dist/api/index.d.ts

@@ -1,5 +1,5 @@
 /**
  * API module exports
  */
-export * from './server.js';
+export * from "./server.js";
 //# sourceMappingURL=index.d.ts.map

package/dist/api/index.js

@@ -1,5 +1,5 @@
 /**
  * API module exports
  */
-export * from './server.js';
+export * from "./server.js";
 //# sourceMappingURL=index.js.map

package/dist/api/server.d.ts

@@ -5,11 +5,14 @@
  *
  * @packageDocumentation
  */
-import { FastifyInstance } from 'fastify';
+import { FastifyInstance } from "fastify";
+import type { IIntentService } from "../intent/index.js";
 /**
  * Create and configure the API server
  */
-export declare function createServer(): Promise<FastifyInstance>;
+export declare function createServer(deps?: {
+    intentService?: IIntentService;
+}): Promise<FastifyInstance>;
 /**
  * Start the API server
  */

package/dist/api/server.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"server.d.ts","sourceRoot":"","sources":["../../src/api/server.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,OAAgB,EAAE,eAAe,EAAgC,MAAM,SAAS,CAAC;AAgLjF;;GAEG;AACH,wBAAsB,YAAY,IAAI,OAAO,CAAC,eAAe,CAAC,CAupB7D;AAED;;GAEG;AACH,wBAAsB,WAAW,IAAI,OAAO,CAAC,IAAI,CAAC,CAsBjD"}
+{"version":3,"file":"server.d.ts","sourceRoot":"","sources":["../../src/api/server.ts"],"names":[],"mappings":"AAAA;;;;;;GAMG;AAEH,OAAgB,EACd,eAAe,EAGhB,MAAM,SAAS,CAAC;AAMjB,OAAO,KAAK,EAAE,cAAc,EAAE,MAAM,oBAAoB,CAAC;AAmLzD;;GAEG;AACH,wBAAsB,YAAY,CAAC,IAAI,CAAC,EAAE;IACxC,aAAa,CAAC,EAAE,cAAc,CAAC;CAChC,GAAG,OAAO,CAAC,eAAe,CAAC,CAgxB3B;AAED;;GAEG;AACH,wBAAsB,WAAW,IAAI,OAAO,CAAC,IAAI,CAAC,CAsBjD"}