nsauditor-ai-agent-skill 0.1.28 → 0.1.30
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 +51 -0
- package/SKILL.md +1 -1
- package/package.json +1 -1
- package/references/plugins.md +1 -0
package/CHANGELOG.md
CHANGED
|
@@ -4,6 +4,57 @@ Release notes for **`nsauditor-ai-agent-skill`** — installable knowledge packa
|
|
|
4
4
|
|
|
5
5
|
---
|
|
6
6
|
|
|
7
|
+
## 0.1.30 — Catalog refresh: EE-RT.21 v2 R2 cleanup for plugin 1024 GCP Cloud Storage Auditor — paired with EE 0.6.9 trio-publish (patch-level R2 reviewer-deferred-items cleanup: Appendix A multi-cloud renderer extension + evidence-gap soc2.json mappings; 5 R1 reviewer folds (0 R-CRITICAL + 1 R-HIGH + 1 R-MEDIUM + 3 R-LOW); plugin count UNCHANGED at 23; 3 new soc2.json mappings; NEW pre-publish doc-consistency gate; twentieth consecutive trio-publish)
|
|
8
|
+
|
|
9
|
+
**Trio-publish institutionalization continued.** Paired with EE 0.6.9 + CE 0.1.63 — **twentieth consecutive trio-publish across EE + CE + agent-skill in a single session** (0.4.5–0.6.9).
|
|
10
|
+
|
|
11
|
+
### What changed
|
|
12
|
+
|
|
13
|
+
- **`references/plugins.md`** — no plugin catalog changes this cycle (plugin count unchanged at 23). Plugin 1024 row from 0.1.29 preserved.
|
|
14
|
+
- **`SKILL.md`** — "post-EE 0.6.8" → "post-EE 0.6.9".
|
|
15
|
+
- **`peerDependencies`** floor: unchanged at `nsauditor-ai >=0.1.40`.
|
|
16
|
+
|
|
17
|
+
### Reviewer-fold highlights (all closed same-session)
|
|
18
|
+
|
|
19
|
+
- **R1-HIGH-1 (mappings)** — Missing C1.1 dual-mapping for `_CAT_METADATA_UNREADABLE` (rationale prose vs JSON-structure drift). Added the parallel C1.1 entry; cross-cloud parity with plugin 1020 S3 precedent restored.
|
|
20
|
+
- **R1-MEDIUM-1 (renderer)** — Strengthened Azure-exclusion comment to cite the engine-projection constraint in addition to plugin-1022 commingling.
|
|
21
|
+
- **R1-LOW-1 (renderer + mappings)** — Cross-control uniqueBuckets dedup test + combined metadata+IAM-failure regression test.
|
|
22
|
+
- **R1-LOW-2 (renderer)** — Narrative phrasing tweak ("AWS S3 / GCS" → "AWS S3 or GCS" for disambiguation).
|
|
23
|
+
|
|
24
|
+
### NEW institutional discipline introduced this cycle
|
|
25
|
+
|
|
26
|
+
**Pre-publish doc-consistency gate** codified in EE's `tasks/CLAUDE.md` after the 0.6.8 → user-caught doc drift (6 stale "22 plugin" claims hid across 4 docs in 2 repos). 22 doc-surface audit checklist + auto-grep + SOC 2 matrix invariant check. Saved as `[[pre_publish_doc_consistency_gate]]` auto-memory for cross-session persistence.
|
|
27
|
+
|
|
28
|
+
### Tests + regression
|
|
29
|
+
|
|
30
|
+
- **EE full regression: 5423/5423 across 851 suites** (was 5415/5415 at 0.6.8 publish; +8 tests, suite count unchanged). **61-session 100% green streak preserved.**
|
|
31
|
+
|
|
32
|
+
---
|
|
33
|
+
|
|
34
|
+
## 0.1.29 — Catalog refresh: NEW plugin 1024 GCP Cloud Storage Auditor — paired with EE 0.6.8 trio-publish (first multi-cloud parity plugin in 6 months; mirrors plugin 1020 AWS S3 Auditor with 6 GCS-specific dimensions; 4 R1 reviewer folds (0 R-CRITICAL + 0 R-HIGH + 3 R-MEDIUM + 1 R-LOW — clean review pass); plugin count 22 → 23; 20 new soc2.json mappings; nineteenth consecutive trio-publish)
|
|
35
|
+
|
|
36
|
+
**Trio-publish institutionalization continued.** Paired with EE 0.6.8 + CE 0.1.62 — **nineteenth consecutive trio-publish across EE + CE + agent-skill in a single session** (0.4.5–0.6.8).
|
|
37
|
+
|
|
38
|
+
### What changed
|
|
39
|
+
|
|
40
|
+
- **`references/plugins.md`** — new plugin 1024 row added (alphanumerically sorted between 1023 zero-trust-checker and 1030 IAM Deep Auditor):
|
|
41
|
+
- **Plugin 1024 — GCP Cloud Storage Auditor** (NEW EE 0.6.8): multi-cloud parity sister of plugin 1020 AWS S3 Auditor. 6 dimensions: bucket-level IAM public bindings (CC6.6 — allUsers = CRITICAL, allAuthenticatedUsers = HIGH), Uniform Bucket-Level Access enforcement (CC6.6 + C1.1 dual-mapped — closes legacy bucket-ACL false-PASS class), Object Versioning (C1.1 + A1.2 dual-mapped), Bucket Lock retention policy (C1.1 + C1.2 dual-mapped; SEC 17a-4 / FINRA 4511 WORM-alignment), Customer-Managed Encryption Keys via Cloud KMS (CC6.1 four-tier custody ladder mirroring plugin 1140 v2 RDS), bucket-level access logging (CC7.1). NEW SDK dep `@google-cloud/storage` in optionalDependencies.
|
|
42
|
+
- **`SKILL.md`** — "post-EE 0.6.7" → "post-EE 0.6.8"; plugin 1024 highlights surfaced; plugin count enumeration 22 → 23.
|
|
43
|
+
- **`peerDependencies`** floor: unchanged at `nsauditor-ai >=0.1.40`.
|
|
44
|
+
|
|
45
|
+
### Reviewer-fold highlights (all closed same-session)
|
|
46
|
+
|
|
47
|
+
- **R-MEDIUM-1** — Severity-ladder co-existence: when both `allUsers` and `allAuthenticatedUsers` are present in different bindings, CRITICAL finding surfaces the HIGH evidence in details + narrative.
|
|
48
|
+
- **R-MEDIUM-2** — Per-bucket runtime exception severity: run()-level catch INFO → LOW for consistency with `_auditBucket` metadata-error pattern.
|
|
49
|
+
- **R-MEDIUM-1 (mappings)** — Cross-cloud parity dual-mappings: 5 soc2.json entries dual-mapped to C1.2 + A1.2 matching AWS S3 precedents.
|
|
50
|
+
- **R-LOW-1** — CMEK regex tightened from substring to full-format 6-segment match.
|
|
51
|
+
|
|
52
|
+
### Tests + regression
|
|
53
|
+
|
|
54
|
+
- **EE full regression: 5415/5415 across 851 suites** (was 5314/5314 at 0.6.7 publish; +101 tests / +17 suites — all attributable to the new plugin file). **60-session 100% green streak preserved.**
|
|
55
|
+
|
|
56
|
+
---
|
|
57
|
+
|
|
7
58
|
## 0.1.28 — Catalog refresh: plugin 1170 v3.1 SG-reference-graph edge dedup + plugin 1200 v6.1 CloudWatch Logs probe retry-on-empty parity — paired with EE 0.6.7 trio-publish (patch-level R2 reviewer-deferred-items cleanup cycle: closes both R2 items from 0.6.6 reviewer pass; 4 R1 reviewer folds (0 R-CRITICAL + 0 R-HIGH + 1 R-MEDIUM + 3 R-LOW — clean review pass) + 1 unanticipated `_retryOnNotFound` two-phase restructure (caught by test interaction); plugin count UNCHANGED at 22; soc2.json UNCHANGED; eighteenth consecutive trio-publish)
|
|
8
59
|
|
|
9
60
|
**Trio-publish institutionalization continued.** Paired with EE 0.6.7 + CE 0.1.61 — **eighteenth consecutive trio-publish across EE + CE + agent-skill in a single session** (0.4.5–0.6.7).
|
package/SKILL.md
CHANGED
|
@@ -297,7 +297,7 @@ CE collision. CE reserves 001-099.
|
|
|
297
297
|
|
|
298
298
|
**Plugin 1170 v3 (EE 0.6.6) SG→SG transitive chain reachability** — `aws-ec2-sg-perimeter-auditor` v3 extension. Pre-v3 each Security Group was audited in isolation; a SG with no direct public-CIDR ingress would emit the PASS-tier "no direct public-internet ingress CIDR rules" finding even if transitively reachable from the internet through a `UserIdGroupPairs` chain. v3 builds the SG-reference graph (`_buildSgReferenceGraph`), identifies public-CIDR roots (`_findPubliclyReachableSgs` — 0.0.0.0/0 / ::/0 ingress), and BFS-walks the graph (`_walkTransitiveReachability`) with cycle defense + depth cap (default 5, max 20) + per-target chain cap (default 10, max 100). 2-hop chains emit **HIGH**; 3+ hop chains emit **CRITICAL** (operator-blindness principle — deeper chains less likely to be noticed). Cross-VPC edges skipped (out-of-scope for v3 v1; INFO trailer). v3 v1 simplification: per-hop port-flow tracked but NOT intersected (`walkthroughRequired=true`). New operator opts: `skipTransitiveReachability` / `transitiveChainDepthCap` / `transitiveChainsPerTargetCap` / `transitiveChainSamplesPerFindingCap`. **v3 R-HIGH-1 fold**: BFS short-circuits enqueue past per-target cap (closes path-enumeration explosion on hub-and-spoke topologies — pre-fold the BFS kept cloning `path` and `visited` Sets and walking past the cap). **v3 R-LOW-2 fold**: depth-cap-hit surfaced separately from per-target-cap (closes silent-deep-truncation false-CLEAN class). 3 new soc2.json mappings under CC6.6 (transitive HIGH + CRITICAL + INFO truncation). **v3.1 EE 0.6.7 closes the edge-dedup R2-deferred item**: `_buildSgReferenceGraph` now dedupes edges by `(sourceGroupId, targetGroupId)` with `ports` aggregated as array of `{protocol, fromPort, toPort}`. Pre-fold a real-world ALB-fronting-app SG with 3 ingress perms on different ports (80/443/8080) referencing the same source SG emitted 3 distinct edges A→B; the BFS treated each as a separate chain, inflating `chainCount` 2-5× and exhausting per-target chain caps on noise. Post-fold the BFS sees exactly 1 chain per distinct (source, target) pair. `isCrossVpc` aggregation is AND-semantic — if ANY contributing pair is same-VPC, the merged edge is same-VPC (per `[[conservative_classifier_principle]]`: walk possibly-same-VPC chains rather than silently skip). Classifier port-render accepts both v3.1 array shape and v3 single-object shape (back-compat). **v3.1 R-MEDIUM-1 fold**: arrival-order independence locked with 2 regression fixtures + JSDoc tightening. **v3.1 R-LOW-1 fold**: partial-render contract on malformed port specs locked with 2 fixtures. **v3.1 R-LOW-2 fold**: `_portKeys` scratch-lifetime documented (MUST NOT escape).
|
|
299
299
|
|
|
300
|
-
**EE SOC 2 substrate-evidence coverage (post-EE 0.6.
|
|
300
|
+
**EE SOC 2 substrate-evidence coverage (post-EE 0.6.9):** 10 covered controls (CC6.1 /
|
|
301
301
|
CC6.2 / CC6.6 / CC6.7 / CC6.8 / CC7.1 / CC7.2 / CC7.3 / C1.1 / C1.2) + 4 partial
|
|
302
302
|
(CC6.3 / CC8.1 / A1.2 / PI1.5) + 33 OOS for static substrate scanning. Coverage matrix
|
|
303
303
|
is institutionally honest: substrate-evidence depth grows release-over-release without
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "nsauditor-ai-agent-skill",
|
|
3
|
-
"version": "0.1.
|
|
3
|
+
"version": "0.1.30",
|
|
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",
|
package/references/plugins.md
CHANGED
|
@@ -169,6 +169,7 @@ listings, and default pages.
|
|
|
169
169
|
| 1021 | GCP Cloud Scanner | Enterprise | Firewall rule audit + IAM bindings + Storage bucket public-access | CC6.1 / CC6.6 / C1.1 |
|
|
170
170
|
| 1022 | Azure Cloud Scanner | Enterprise | NSG rule analysis + RBAC role assignments + Storage account hardening | CC6.1 / CC6.6 / C1.1 |
|
|
171
171
|
| 1023 | Zero Trust Checker | Enterprise | Segmentation + encryption + identity + lateral movement scoring (reads OBSERVED open ports) | CC6.6 |
|
|
172
|
+
| 1024 | GCP Cloud Storage Auditor (NEW EE 0.6.8) | Enterprise | Multi-cloud parity sister of plugin 1020 AWS S3. 6 dimensions: bucket-level IAM public bindings (allUsers = CRITICAL, allAuthenticatedUsers = HIGH); Uniform Bucket-Level Access enforcement (closes legacy bucket-ACL false-PASS class); Object Versioning; Bucket Lock retention policy (SEC 17a-4 / FINRA 4511 WORM-alignment); CMEK via Cloud KMS (four-tier custody ladder); bucket-level access logging. NEW SDK dep `@google-cloud/storage`. | CC6.1 / CC6.6 / CC7.1 / C1.1 / C1.2 / A1.2 |
|
|
172
173
|
| 1030 | AWS IAM Deep Auditor | Enterprise | Shadow-admin path detection via BFS over PassRole / AssumeRole / federated trust; restrictive-Condition allowlist (Auth0 / Okta / Cognito) | CC6.1 |
|
|
173
174
|
| 1040 | AWS CloudTrail Operational Integrity | Enterprise | Trail health + log-file validation + KMS-CMK; CloudWatch alarm coverage vs CIS AWS Foundations Benchmark v1.5 §3.1–3.14; cross-account S3 trail-destination WORM verification (SEC 17a-4 / FINRA 4511) | CC7.2 / CC7.3 |
|
|
174
175
|
| 1050 | AWS API Gateway Assurance | Enterprise | Per-method/route authorization classifier; custom-domain TLS policy; stage-level access logging / throttling / WAF; public-endpoint exposure; Lambda authorizer cross-reference via lambda:GetFunction | CC6.1 / CC6.6 / CC6.7 / CC7.1 / A1.2 |
|