swt3-ai 0.7.2 → 0.7.4
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/README.md +44 -1
- metadata +2 -2
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: c85be5fb87d1fc3299f6d1822c28a17878ae56f126629f07c147856346f7aacf
|
|
4
|
+
data.tar.gz: 80a0612bac2f25b12542cc12288be63acf316a44fcdee46239718c653b87b9b3
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: be63b6e8f38b9b55be9caee367c02975ed116197b8693541c50271008bdb37ec975c68b31f068561fe5ab4fc540e4c49b6ed32b693e09166d4eb2e6cb17173da
|
|
7
|
+
data.tar.gz: 8eaa1f797feb78777abb99159b9f20f3ed0f2b2c94c20929ac2ce7d0739c4cc2773580c347a0829d002727657ec0c21e4d3c93b8e6591252d45e7576604538a7
|
data/README.md
CHANGED
|
@@ -11,6 +11,49 @@ Built for platform engineers who need to bolt compliance witnessing onto product
|
|
|
11
11
|
|
|
12
12
|
EU AI Act GPAI transparency obligations enforced since **August 2, 2026**. High-risk enforcement follows **December 2, 2027**. This SDK gives you the cryptographic primitives for both.
|
|
13
13
|
|
|
14
|
+
> **Protocol Spec:** [swt3.ai/spec](https://swt3.ai/spec) | **Registry:** [swt3.ai/registry](https://swt3.ai/registry) | **Verify:** [swt3.ai/verify](https://swt3.ai/verify)
|
|
15
|
+
|
|
16
|
+
## What's New in v0.7.4
|
|
17
|
+
|
|
18
|
+
NVIDIA shipped OpenShell to 120+ partners. Agent containment is now infrastructure. But containment without evidence is a black box -- an auditor cannot verify that a sandbox policy was enforced last Tuesday at 14:00 UTC by looking at enforcement logs alone. v0.7.4 adds runtime containment attestation (AI-SHELL.1), model distillation provenance (AI-DIST.1), MCP elicitation detection (AI-MCP.6), and incident lifecycle chains. Four governance gaps closed in one release.
|
|
19
|
+
|
|
20
|
+
**Why this matters for Ruby:** All three new procedure IDs (AI-SHELL.1, AI-DIST.1, AI-MCP.6) verify with the existing `mint_fingerprint` function. AI-SHELL.1 runtime containment attestation works with zero dependency changes -- Ruby API backends orchestrating containerized inference pipelines can attest containment policy enforcement with the same fingerprint output as every other SDK language.
|
|
21
|
+
|
|
22
|
+
### 3 New Procedures
|
|
23
|
+
|
|
24
|
+
- **AI-SHELL.1** (Runtime Containment Attestation): Records that a sandboxed runtime enforced its containment policy during a specific observation window. Duck-typed: works with OpenShell, gVisor, Kata, Firecracker, WASM. [NVIDIA OpenShell Crosswalk](https://sovereign.tenova.io/guides/nvidia-openshell-crosswalk.html)
|
|
25
|
+
- **AI-DIST.1** (Distillation Provenance): Records teacher-to-student model lineage, compression ratio, and knowledge transfer method.
|
|
26
|
+
- **AI-MCP.6** (Elicitation Detection): Records prompt injection attempts via MCP tool responses. OWASP MCP-06.
|
|
27
|
+
|
|
28
|
+
### Updated Coverage
|
|
29
|
+
|
|
30
|
+
- 278 procedures across 10 namespaces (was 284)
|
|
31
|
+
- 75 MCP tools (was 72)
|
|
32
|
+
- 277 compliance guides
|
|
33
|
+
- [NVIDIA OpenShell Crosswalk](https://sovereign.tenova.io/guides/nvidia-openshell-crosswalk.html) -- runtime containment evidence for OpenShell, gVisor, Kata, Firecracker
|
|
34
|
+
- All new anchor types verify with `Swt3.mint_fingerprint()` -- no library update required for verification
|
|
35
|
+
|
|
36
|
+
## What's New in v0.7.3
|
|
37
|
+
|
|
38
|
+
A2A (Google's Agent-to-Agent protocol) has 150+ supporting organizations, all three hyperscalers, and a v1.0 stable specification under Linux Foundation governance. It has zero built-in audit trail. MCP OAuth adoption sits at 8.5% with 30+ CVEs filed in 60 days. v0.7.3 makes SWT3 the evidence layer for both agent communication protocols with dedicated procedures, lifecycle-aware adapters, and a new crosswalk CLI command.
|
|
39
|
+
|
|
40
|
+
**Why this matters for Ruby:** Ruby API backends and Rails agent orchestrators gain A2A lifecycle attestation with the same factor semantics and fingerprint parity as every other SDK language.
|
|
41
|
+
|
|
42
|
+
### 4 New Procedures
|
|
43
|
+
|
|
44
|
+
- **AI-A2A.1** (Task Delegation Lifecycle): Records task state transitions (submitted/working/input_required/completed/failed/canceled/rejected) with delegation depth and latency.
|
|
45
|
+
- **AI-A2A.2** (Agent Card Discovery): Records agent discovery via well-known URLs, registries, or referrals, with verified credential count.
|
|
46
|
+
- **AI-A2A.3** (Context Chain Linking): Records contextId linkage across multi-agent delegation chains for forensic reconstruction.
|
|
47
|
+
- **AI-MCP.5** (OAuth Token Binding): Records 8 OAuth lifecycle events (discovery through revocation) with binding strength and scope governance.
|
|
48
|
+
|
|
49
|
+
### Updated Coverage
|
|
50
|
+
|
|
51
|
+
- 284 procedures across 10 namespaces (was 280/9)
|
|
52
|
+
- 72 MCP tools (was 68)
|
|
53
|
+
- A2A namespace added (10th namespace)
|
|
54
|
+
- `swt3 crosswalk <procedure>` CLI command for offline framework mapping lookup
|
|
55
|
+
- 269 compliance guides
|
|
56
|
+
|
|
14
57
|
## What's New in v0.7.2
|
|
15
58
|
|
|
16
59
|
Two gaps closed in one release: harness-layer governance and zero-code MCP compliance. v0.7.2 adds five new procedures for orchestration, context management, sandboxing, and eval gating -- the decisions AI harnesses make before inference that have no audit trail today. It also ships a Witness Middleware (`withSWT3(transport)`) in the companion MCP package that auto-mints AI-TOOL.1 anchors for every tool call flowing through any MCP server, with zero code changes to tool handlers.
|
|
@@ -299,7 +342,7 @@ Your prompts and responses **never leave your infrastructure**. The SDK computes
|
|
|
299
342
|
## Links
|
|
300
343
|
|
|
301
344
|
- **Website**: [tenova.io](https://tenova.io)
|
|
302
|
-
- **Protocol Spec**: [SWT3
|
|
345
|
+
- **Protocol Spec**: [SWT3 Protocol Specification](https://swt3.ai/spec)
|
|
303
346
|
- **Live Demo**: [sovereign.tenova.io/audit/axm_audit_demo_eu_ai_act_public](https://sovereign.tenova.io/audit/axm_audit_demo_eu_ai_act_public)
|
|
304
347
|
|
|
305
348
|
---
|
metadata
CHANGED
|
@@ -1,14 +1,14 @@
|
|
|
1
1
|
--- !ruby/object:Gem::Specification
|
|
2
2
|
name: swt3-ai
|
|
3
3
|
version: !ruby/object:Gem::Version
|
|
4
|
-
version: 0.7.
|
|
4
|
+
version: 0.7.4
|
|
5
5
|
platform: ruby
|
|
6
6
|
authors:
|
|
7
7
|
- TeNova Labs
|
|
8
8
|
autorequire:
|
|
9
9
|
bindir: bin
|
|
10
10
|
cert_chain: []
|
|
11
|
-
date: 2026-09-
|
|
11
|
+
date: 2026-09-29 00:00:00.000000000 Z
|
|
12
12
|
dependencies: []
|
|
13
13
|
description: Mint, verify, and sign SWT3 witness anchors for AI compliance. EU AI
|
|
14
14
|
Act, NIST AI RMF, CMMC, SR 11-7. Cross-language parity with Python and TypeScript
|