nsauditor-ai-agent-skill 0.2.25 → 0.2.26

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.
package/CHANGELOG.md CHANGED
@@ -4,6 +4,12 @@ Release notes for **`nsauditor-ai-agent-skill`** — installable knowledge packa
4
4
 
5
5
  ---
6
6
 
7
+ ## 0.2.26 (2026-07-12) — Paired content bump for EE 0.32.3 (RDS cluster-level SSL enforcement + CE GRC-push preflight)
8
+
9
+ Paired content bump for the EE 0.32.3 trio. SKILL.md reflects the RDS auditor's new **cluster-level SSL enforcement** dimension (plugin 1140 now audits the Aurora **cluster** parameter group via `DescribeDBClusterParameters` — closing a cleartext false-negative on instance-less Aurora Serverless v1 clusters that the instance-level SSL check never saw) and the **staged-parameter (`ParameterApplyStatus`) discipline** (a set-but-not-yet-applied `rds.force_ssl` / `pgaudit` is no longer affirmed as effective). Also notes the CE CLI's new **GRC-push startup preflight** (fail-fast on bad GRC config before the scan; CLI-only, not MCP-reachable). Matrix-neutral: no new framework, plugin count UNCHANGED at 28, all seven coverage matrices UNCHANGED. Paired **EE 0.32.3** + **CE 0.2.28**.
10
+
11
+ ---
12
+
7
13
  ## 0.2.25 (2026-07-10) — Paired content bump for EE 0.32.2 (GRC connector trio complete — Secureframe + cross-framework report-quality leak closure)
8
14
 
9
15
  Paired content bump for the EE 0.32.2 trio. SKILL.md now teaches the completed **Vanta · Drata · Secureframe** GRC connector trio (enable Secureframe with `COMPLIANCE_GRC_PROVIDER=secureframe` — records model, your Secureframe rules evaluate the pushed control records; outbound, single-workspace, opt-in early-access; API shape published-assumed, live-tenant validation deferred) and reflects the closed **cross-framework foreign-token leak** in the "Why this violates" rationales (no internal `Inherits from soc2.json` note, bare foreign control-ids, cross-framework routing-maps, or `real-engine verified ==` QA-note leaks a foreign framework's name into a customer's Report on Compliance). Matrix-neutral: no new framework, plugin count UNCHANGED at 28, all seven coverage matrices UNCHANGED. Paired **EE 0.32.2** + **CE 0.2.27**.
package/SKILL.md CHANGED
@@ -16,7 +16,7 @@ description: >
16
16
 
17
17
  # NSAuditor AI — Agent Skill
18
18
 
19
- > **Version:** 0.2.25 (post-EE-0.32.2 — a **matrix-neutral** release: **no new framework, no new plugins (still 28), all seven coverage matrices UNCHANGED.** Two legs: (1) a new **Secureframe** GRC push connector completes the **Vanta · Drata · Secureframe** trio at the same early-access opt-in shape — set `COMPLIANCE_GRC_PROVIDER=secureframe` + `COMPLIANCE_GRC_TOKEN` and the compliance phase pushes structured control records to your Secureframe workspace evidence collection (records model like Drata — your Secureframe rules evaluate them; the connector carries the control `status` verbatim, it does NOT compute pass/fail; outbound, single-workspace, opt-in early-access; API shape published-assumed, live-tenant validation deferred to partner intake); and (2) the **cross-framework foreign-token leak** in the "Why this violates" rationales is closed — an internal `Inherits from soc2.json CC6.1` note, bare foreign control-ids, cross-framework routing-maps, and a `real-engine verified ==` QA-note no longer leak a **foreign** framework's name into a HIPAA / PCI / ISO / NIST / CIS / GDPR Report on Compliance (~300 rationales across all seven frameworks; pure-deletion, subsequence-invariant-proven, with a class-level guard; routing byte-neutral). Prior context — EE 0.32.1 cleaned internal engineering markers (`[[wiki-links]]`, `EE-RT` work-codes, reviewer-codes, `Rn-SEVERITY` review-round IDs) out of the rationales + deeper Azure/AWS positive-substrate + a Vanta/Drata push-loop DRY refactor; EE 0.32.0 wired scan-time GRC push (Vanta) + shipped the Drata connector. Honest status: the three connectors are shipped + opt-in + extensively tested; live validation against production tenants in progress — early-access, single-workspace, not a multi-tenant sync. The seven frameworks: SOC 2 · HIPAA · NIST CSF 2.0 · PCI DSS v4.0.1 · ISO/IEC 27001:2022 · CIS Controls v8 · **GDPR Article 32** — the last is **Art. 32 infrastructure substrate ONLY, NOT GDPR compliance** [Art. 32 is the only one of GDPR's 99 articles an infrastructure scanner can evidence; the rest is operator-side, out of scope by design], carrying four-factor proportionality + personal-data-scope attestation + the Art. 83(4) LOWER fine tier [€10M/2%, not the €20M/4% headline tier]. Paired CE 0.2.27 + EE 0.32.2) · **Source:** [github.com/nsasoft/nsauditor-ai](https://github.com/nsasoft/nsauditor-ai) · **npm:** `nsauditor-ai` · **License:** MIT (CE)
19
+ > **Version:** 0.2.26 (post-EE-0.32.3 — a **matrix-neutral** RDS false-negative depth pass: **no new framework, no new plugins (still 28), all seven coverage matrices UNCHANGED.** Plugin 1140 now audits Aurora **cluster-level SSL enforcement** (`DescribeDBClusterParameters`), closing a cleartext false-negative on **instance-less Aurora Serverless v1** clusters the instance-level SSL check never saw (a cluster whose cluster parameter group set `rds.force_ssl=0` / `require_secure_transport=OFF` produced zero transit-encryption findings → now CRITICAL), and applies **staged-parameter (`ParameterApplyStatus`) discipline** so a set-but-not-yet-applied `rds.force_ssl` / `pgaudit` is no longer affirmed as effective. The CE CLI adds a **GRC-push startup preflight** — fail-fast on a bad GRC token / control-map / provider at scan startup before the full scan (CLI-only, not MCP-reachable). The new cluster findings route to the existing transit-encryption controls (titlePattern-only anchor edit). Prior context — EE 0.32.2 completed the **Vanta · Drata · Secureframe** GRC connector trio (records model, opt-in early-access, outbound single-workspace, live-tenant validation deferred to partner intake) + closed the cross-framework foreign-token leak in the "Why this violates" rationales; EE 0.32.1 cleaned internal engineering markers out of the rationales. Honest status: the three connectors are shipped + opt-in + extensively tested; live validation against production tenants in progress — early-access, single-workspace, not a multi-tenant sync. The seven frameworks: SOC 2 · HIPAA · NIST CSF 2.0 · PCI DSS v4.0.1 · ISO/IEC 27001:2022 · CIS Controls v8 · **GDPR Article 32** — the last is **Art. 32 infrastructure substrate ONLY, NOT GDPR compliance** [Art. 32 is the only one of GDPR's 99 articles an infrastructure scanner can evidence; the rest is operator-side, out of scope by design], carrying four-factor proportionality + personal-data-scope attestation + the Art. 83(4) LOWER fine tier [€10M/2%, not the €20M/4% headline tier]. Paired CE 0.2.28 + EE 0.32.3) · **Source:** [github.com/nsasoft/nsauditor-ai](https://github.com/nsasoft/nsauditor-ai) · **npm:** `nsauditor-ai` · **License:** MIT (CE)
20
20
 
21
21
  NSAuditor AI is a modular, AI-assisted network security audit platform with 27+ scanner
22
22
  plugins, CVE matching, MITRE ATT&CK mapping, and Zero Data Exfiltration by design. This
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "nsauditor-ai-agent-skill",
3
- "version": "0.2.25",
3
+ "version": "0.2.26",
4
4
  "description": "AI Agent Skill for NSAuditor AI — gives any AI coding agent built-in knowledge of NSAuditor's MCP tools, schemas, plugins, and security audit workflows.",
5
5
  "keywords": [
6
6
  "nsauditor",