nsauditor-ai-agent-skill 0.2.34 → 0.2.36
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 +55 -14
- package/README.md +1 -1
- package/SKILL.md +15 -3
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -4,23 +4,64 @@ Release notes for **`nsauditor-ai-agent-skill`** — installable knowledge packa
|
|
|
4
4
|
|
|
5
5
|
---
|
|
6
6
|
|
|
7
|
+
## 0.2.36 (2026-08-07) — the opt-in RFC 3161 wording corrected: a hedge that had outlived its measurement
|
|
8
|
+
|
|
9
|
+
`SKILL.md` described RFC 3161 timestamping as *"implemented but not yet wired to a flag — do
|
|
10
|
+
not quote it as shipping"*. EE 0.33.0 wired it via `NSAUDITOR_TSA_URL`, and on 2026-08-07 the
|
|
11
|
+
live smoke exercised it against a real Time-Stamp Authority through the published binary
|
|
12
|
+
(`openssl ts -verify` returning OK, with a one-byte-mutated copy returning FAILED as the
|
|
13
|
+
control). The sentence had been false for a full release, and because this file is
|
|
14
|
+
instruction material for AI agents it did not merely sit there — a live Claude Desktop reply
|
|
15
|
+
repeated it downstream, which is how it was found.
|
|
16
|
+
|
|
17
|
+
The row now states what is true: opt-in via `NSAUDITOR_TSA_URL`, no CLI flag and no default,
|
|
18
|
+
an outbound call to the Time-Stamp Authority you name, and exercised against a live authority
|
|
19
|
+
on the npm path. ⚠️ **Superseded within the day:** the in-container live round-trip was then
|
|
20
|
+
run too, from inside the pushed `:0.33.0` image, so BOTH delivery vehicles are proven end to
|
|
21
|
+
end. The only container caveat that survives is version scope — retained images at `:0.32.11`
|
|
22
|
+
and earlier carry no `openssl` and nothing here speaks for them.
|
|
23
|
+
|
|
24
|
+
No behaviour change; wording only. Paired with the EE-side renderer correction on the same
|
|
25
|
+
root cause.
|
|
26
|
+
|
|
27
|
+
---
|
|
28
|
+
|
|
29
|
+
## 0.2.35 (2026-08-07) — paired with EE 0.33.0 · requires CE ≥ 0.2.37
|
|
30
|
+
|
|
31
|
+
Three new CE entry points to teach: `compliance attest` (Type II recurring-scan attestation;
|
|
32
|
+
exit 3 on an empty history, because absence of evidence is the finding), plus `--sla-policy`
|
|
33
|
+
and `--compliance-history`. A startup posture veto refuses an offline-plus-egress
|
|
34
|
+
contradiction with exit 2.
|
|
35
|
+
|
|
36
|
+
Two corrections to this package's own prose:
|
|
37
|
+
|
|
38
|
+
- The `compliance_check` disclosure in `SKILL.md` now carries the **WITHDRAWN** marker in the
|
|
39
|
+
same sentence. An absence guard cannot tell a disclosure from the thing it discloses, and
|
|
40
|
+
the sentence whose whole job is to say the tool does not exist was being read as a claim.
|
|
41
|
+
- The Enterprise row no longer calls plugin **1023** a cloud auditor. Measured by driving the
|
|
42
|
+
CLI: it scores zero-trust posture from a **network-host** scan, never runs on a cloud pass,
|
|
43
|
+
and cannot be invoked with `--plugins 1023` — it needs a discovery plugin to confirm the
|
|
44
|
+
host is up first, so selecting it alone removes its own precondition.
|
|
45
|
+
|
|
46
|
+
Plugin catalog UNCHANGED at 28 EE / 55 overall; all seven matrices UNCHANGED.
|
|
47
|
+
|
|
7
48
|
## 0.2.34 (2026-08-05) — Paired with EE 0.32.11 · a withdrawn capability, and SEVEN MCP tools that do not exist
|
|
8
49
|
|
|
9
50
|
### The tools that were never there
|
|
10
51
|
|
|
11
|
-
Found by Gate-3 prompt 6, in a live reply. Asked how to build a SOC 2 Type II evidence package, the assistant answered: *"`scan_compare` (Pro) gives risk-weighted deltas between runs, which is how you show a finding appeared, was remediated, and stayed closed."*
|
|
52
|
+
Found by Gate-3 prompt 6, in a live reply. Asked how to build a SOC 2 Type II evidence package, the assistant answered: *"`scan_compare` (Pro) gives risk-weighted deltas between runs, which is how you show a finding appeared, was remediated, and stayed closed."* *(WITHDRAWN — recorded here as the withdrawal itself.)*
|
|
12
53
|
|
|
13
|
-
**`scan_compare` does not exist.** Measured against the live registry, SEVEN tools this package advertised return 0 entries in `TOOLS` and `undefined` from `toolHandlers`:
|
|
54
|
+
**`scan_compare` is WITHDRAWN — it does not exist.** Measured against the live registry, SEVEN tools this package advertised return 0 entries in `TOOLS` and `undefined` from `toolHandlers`:
|
|
14
55
|
|
|
15
|
-
`risk_summary` · `scan_compare` · `save_finding` · `start_assessment` · `prioritize_risks` · `compliance_check` · `export_report`
|
|
56
|
+
WITHDRAWN, all seven: `risk_summary` · `scan_compare` · `save_finding` · `start_assessment` · `prioritize_risks` · `compliance_check` · `export_report`
|
|
16
57
|
|
|
17
58
|
The shipped set is exactly `scan_host`, `scan_cloud`, `get_findings`, `compliance_matrix`, `probe_service`, `get_vulnerabilities`, `list_plugins`.
|
|
18
59
|
|
|
19
|
-
**And the seven are not one class — a correction to this entry's first draft.** Four (`compliance_check`, `export_report`, `start_assessment`, `prioritize_risks`) have **no implementation anywhere**: `registerEnterpriseTools()` is an empty no-op. Three (`risk_summary`, `scan_compare`, `save_finding`) are **implemented but UNREACHABLE** — real handlers and real tests exist in EE's `PRO_MCP_TOOLS`, but `registerProTools()` has **zero non-test call sites**, so nothing ever registers them. Either way the skill must not advertise them, but the fixes differ: the first four need building or forgetting; the last three need WIRING, and are on the EE board as such.
|
|
60
|
+
**And the seven are not one class — a correction to this entry's first draft.** Four (`compliance_check`, `export_report`, `start_assessment`, `prioritize_risks`) have **no implementation anywhere**: `registerEnterpriseTools()` is an empty no-op. Three (`risk_summary`, `scan_compare`, `save_finding`) are **implemented but UNREACHABLE** — real handlers and real tests exist in EE's `PRO_MCP_TOOLS`, but `registerProTools()` has **zero non-test call sites**, so nothing ever registers them. Either way the skill must not advertise them, but the fixes differ: the first four need building or forgetting; the last three need WIRING, and are on the EE board as such. *(WITHDRAWN — recorded here as the withdrawal itself.)*
|
|
20
61
|
|
|
21
|
-
`compliance_check` was the worst: it appeared **4 times in `SKILL.md` and 13 times across the package**, including a behavioural imperative — *"If unsure, run `compliance_check` and report what the pack says."* *(An earlier draft of this entry said "11 occurrences"; that figure was wrong and is corrected here — counted, not remembered.)* An agent loads this file into context and follows its imperatives, so an operator assembling the artifact an auditor SAMPLES was being routed at a tool that is not there. `export_report` promised "PDF, HTML" — the same phantom as the `pdfExport` capability flag EE
|
|
62
|
+
`compliance_check` was the worst: it appeared **4 times in `SKILL.md` and 13 times across the package**, including a behavioural imperative — *"If unsure, run `compliance_check` and report what the pack says."* *(An earlier draft of this entry said "11 occurrences"; that figure was wrong and is corrected here — counted, not remembered.)* An agent loads this file into context and follows its imperatives, so an operator assembling the artifact an auditor SAMPLES was being routed at a tool that is not there. `export_report` promised "PDF, HTML" — the same phantom as the `pdfExport` capability flag, WITHDRAWN by EE this cycle, wearing a different name. *(WITHDRAWN — recorded here as the withdrawal itself.)*
|
|
22
63
|
|
|
23
|
-
**The correction is not a deletion.** `compliance_check`'s row carried a great deal of accurate framework detail; what was wrong was the MECHANISM it hung on. Compliance is a **CLI** surface: `nsauditor-ai scan --host <target> --compliance <fw> --out <dir>` runs the compliance phase and writes the pack. `scan_cloud` maps findings to frameworks but never runs that phase, so **no MCP call produces a pack**. The detail is kept and re-attributed; use `compliance_matrix` to state COVERAGE and the CLI to produce EVIDENCE.
|
|
64
|
+
**The correction is not a deletion.** `compliance_check`'s row carried a great deal of accurate framework detail; what was wrong was the MECHANISM it hung on. Compliance is a **CLI** surface: `nsauditor-ai scan --host <target> --compliance <fw> --out <dir>` runs the compliance phase and writes the pack. `scan_cloud` maps findings to frameworks but never runs that phase, so **no MCP call produces a pack**. The detail is kept and re-attributed; use `compliance_matrix` to state COVERAGE and the CLI to produce EVIDENCE. *(WITHDRAWN — recorded here as the withdrawal itself.)*
|
|
24
65
|
|
|
25
66
|
**`compliance_matrix` is now documented** — it shipped in CE 0.2.36 and this package had never mentioned it, which is the other direction of the same defect: a tool an agent is not told about is a tool it will not reach for.
|
|
26
67
|
|
|
@@ -28,7 +69,7 @@ The shipped set is exactly `scan_host`, `scan_cloud`, `get_findings`, `complianc
|
|
|
28
69
|
|
|
29
70
|
`gate:claims` sweeps for withdrawn capability WORDING. **A tool NAME matches no claim pattern** — it is a claim that exists only as an identifier, exactly like the withdrawn capability flags. EE now ships `tests/agent_skill_tool_surface.test.mjs`, which pins this package's advertised tool set against CE's live `TOOLS` **in both directions** (advertised-but-absent, and shipped-but-undocumented), fails rather than skips when the sibling repo is missing, and proves its own extractor on a probe. Both directions are mutation-proven.
|
|
30
71
|
|
|
31
|
-
> The sweep that first found this was itself too narrow — it enumerated the name prefixes already seen and so missed `start_assessment` and `prioritize_risks`. The guard therefore extracts the documentation STRUCTURALLY rather than grepping for tool-shaped words.
|
|
72
|
+
> The sweep that first found this was itself too narrow — it enumerated the name prefixes already seen and so missed `start_assessment` and `prioritize_risks`. The guard therefore extracts the documentation STRUCTURALLY rather than grepping for tool-shaped words. *(WITHDRAWN — recorded here as the withdrawal itself.)*
|
|
32
73
|
|
|
33
74
|
---
|
|
34
75
|
|
|
@@ -113,7 +154,7 @@ Paired version bump for the EE 0.32.1 trio — matrix-neutral patch. EE 0.32.1 c
|
|
|
113
154
|
|
|
114
155
|
## 0.2.23 (2026-07-07) — Paired content bump for EE 0.32.0 (Enterprise GRC push activation — Vanta + Drata)
|
|
115
156
|
|
|
116
|
-
Paired content bump (SKILL.md version line + the `compliance_check` capability's GRC-connector mention + the `COMPLIANCE_GRC_PROVIDER` Environment-Variables entry + this changelog). **EE 0.32.0** wires the dormant Vanta connector for **scan-time GRC push activation** (`COMPLIANCE_GRC_PROVIDER=vanta` + `COMPLIANCE_GRC_TOKEN` → the compliance phase maps findings to Vanta test results and pushes them, opt-in + ZDE-redacted, token never serialized) and ships a new **Drata** connector (Custom Connections; your Drata Test Builder rules evaluate the pushed records); Secureframe is on the roadmap. Also ships the **T1/T2 AWS positive-substrate curation** (60 PASS-tier findings opt into the per-control RoC positive-substrate view; display-only, non-flipping, count-neutral). Honest status: connectors shipped + opt-in + extensively tested; live validation against production tenants is in progress — early-access, single-workspace, not a multi-tenant sync. **No new MCP tools or schemas. No new framework, no new plugins (still 28), all seven coverage matrices UNCHANGED.** Paired **EE 0.32.0** + **CE 0.2.25**.
|
|
157
|
+
Paired content bump (SKILL.md version line + the `compliance_check` capability's GRC-connector mention + the `COMPLIANCE_GRC_PROVIDER` Environment-Variables entry + this changelog). **EE 0.32.0** wires the dormant Vanta connector for **scan-time GRC push activation** (`COMPLIANCE_GRC_PROVIDER=vanta` + `COMPLIANCE_GRC_TOKEN` → the compliance phase maps findings to Vanta test results and pushes them, opt-in + ZDE-redacted, token never serialized) and ships a new **Drata** connector (Custom Connections; your Drata Test Builder rules evaluate the pushed records); Secureframe is on the roadmap. Also ships the **T1/T2 AWS positive-substrate curation** (60 PASS-tier findings opt into the per-control RoC positive-substrate view; display-only, non-flipping, count-neutral). Honest status: connectors shipped + opt-in + extensively tested; live validation against production tenants is in progress — early-access, single-workspace, not a multi-tenant sync. **No new MCP tools or schemas. No new framework, no new plugins (still 28), all seven coverage matrices UNCHANGED.** Paired **EE 0.32.0** + **CE 0.2.25**. *(WITHDRAWN at 0.2.34 — the tool named here does not exist; this line is the historical record of what the package said at the time.)*
|
|
117
158
|
|
|
118
159
|
## 0.2.22 (2026-07-05) — Paired content bump for the CE 0.2.24 multi-cloud scope-integrity fix + EE 0.31.10 (GCP gate + T4 positive-substrate)
|
|
119
160
|
|
|
@@ -141,7 +182,7 @@ Paired content bump for the EE 0.31.7 cycle (SKILL.md version line + this change
|
|
|
141
182
|
|
|
142
183
|
## 0.2.18 (2026-06-27) — Paired content bump for EE 0.31.6 (RDS enumeration-truncation no-false-clean class CLOSED + audit-log retention routing-depth sweep; CIS matrix 17/22/114 → 17/23/113)
|
|
143
184
|
|
|
144
|
-
Paired content bump for the EE 0.31.6 cycle (SKILL.md version line + the `compliance_check` CIS count). EE 0.31.6 makes every RDS enumerator fail closed on page-cap truncation, registers the RDS auditor in the compliance-engine drift detector, and maps RDS audit-log retention to PCI DSS 10.5.1 / NIST CSF PR.PS-04 / ISO A.8.15 / CIS Safeguard 8.10 — flipping **CIS Controls v8 Safeguard 8.10 OOS → partial** (CIS matrix **17/22/114 → 17/23/113**; IG1 cyber-insurance baseline UNCHANGED at 23/56, IG2-cum 38, IG3-cum 40). The other six matrices are UNCHANGED; plugin count UNCHANGED at 28. Paired **EE 0.31.6** + CE 0.2.20.
|
|
185
|
+
Paired content bump for the EE 0.31.6 cycle (SKILL.md version line + the `compliance_check` CIS count). EE 0.31.6 makes every RDS enumerator fail closed on page-cap truncation, registers the RDS auditor in the compliance-engine drift detector, and maps RDS audit-log retention to PCI DSS 10.5.1 / NIST CSF PR.PS-04 / ISO A.8.15 / CIS Safeguard 8.10 — flipping **CIS Controls v8 Safeguard 8.10 OOS → partial** (CIS matrix **17/22/114 → 17/23/113**; IG1 cyber-insurance baseline UNCHANGED at 23/56, IG2-cum 38, IG3-cum 40). The other six matrices are UNCHANGED; plugin count UNCHANGED at 28. Paired **EE 0.31.6** + CE 0.2.20. *(WITHDRAWN at 0.2.34 — the tool named here does not exist; this line is the historical record of what the package said at the time.)*
|
|
145
186
|
|
|
146
187
|
---
|
|
147
188
|
|
|
@@ -151,7 +192,7 @@ Paired content bump — no CE engine behavior change (detection + routing live i
|
|
|
151
192
|
|
|
152
193
|
## 0.2.16 (2026-06-25) — Paired content bump for EE 0.31.4 (cloud-scan presentation false-clean fix + `--compliance all` / fail-fast validation)
|
|
153
194
|
|
|
154
|
-
Paired content bump — no new framework, no new plugins (still 28), all seven coverage matrices UNCHANGED. EE 0.31.4 is a report-surface + CLI-UX hardening patch: a cloud scan (`--host aws|azure|gcp`) with real findings no longer surfaces the CE network concluder's *"Host is UP — No open services detected"* headline (the conclusion is rewritten to a cloud-appropriate N-by-severity summary with top risks), and a plugin that **times out or errors** routes to **coverage UNVERIFIED** rather than an affirmative clean verdict — the cardinal presentation-layer false-clean, closed (detection unchanged, oracle-validated). `--compliance all` expands to all seven frameworks; an unknown token **fails fast** (no "Framework load failed" stub). The skill's `compliance_check` tool reference + HIPAA usage example now document `--compliance all`; the version blurb is refreshed. Paired with EE 0.31.4 + CE 0.2.16. **EE 0.31.4 requires CE 0.2.8+.**
|
|
195
|
+
Paired content bump — no new framework, no new plugins (still 28), all seven coverage matrices UNCHANGED. EE 0.31.4 is a report-surface + CLI-UX hardening patch: a cloud scan (`--host aws|azure|gcp`) with real findings no longer surfaces the CE network concluder's *"Host is UP — No open services detected"* headline (the conclusion is rewritten to a cloud-appropriate N-by-severity summary with top risks), and a plugin that **times out or errors** routes to **coverage UNVERIFIED** rather than an affirmative clean verdict — the cardinal presentation-layer false-clean, closed (detection unchanged, oracle-validated). `--compliance all` expands to all seven frameworks; an unknown token **fails fast** (no "Framework load failed" stub). The skill's `compliance_check` tool reference + HIPAA usage example now document `--compliance all`; the version blurb is refreshed. Paired with EE 0.31.4 + CE 0.2.16. **EE 0.31.4 requires CE 0.2.8+.** *(WITHDRAWN at 0.2.34 — the tool named here does not exist; this line is the historical record of what the package said at the time.)*
|
|
155
196
|
|
|
156
197
|
## 0.2.15 (2026-06-24) — Paired content bump for EE 0.31.3 (Enumeration-failure fleet sweep + Aurora DB-cluster snapshot dimension)
|
|
157
198
|
|
|
@@ -167,7 +208,7 @@ Paired content bump — no new framework. EE 0.30.1 closes the last two AWS sour
|
|
|
167
208
|
|
|
168
209
|
## 0.2.12 (2026-06-18) — Paired content bump for EE 0.30.0 (AWS + Azure false-negative depth-pass + cross-source compliance-mapping parity)
|
|
169
210
|
|
|
170
|
-
Paired content bump — no new framework. EE 0.30.0 is a detection-depth + mapping-correctness release: it closes a wave of cloud-misconfiguration false-negatives across AWS (S3 access-points · resource-policy effective-exposure · EC2-SG public-CIDR/split-range · KMS effective-decrypt/cross-account · IAM ListUsers truncation) and Azure (storage/NSG/Key-Vault/cloud-scanner class-O fail-opens), and routes architecturally-identical confidentiality / least-privilege exposures consistently across all seven frameworks (e.g. a public application entry-point now appears on a NIST CSF **PR.AA-05** least-privilege report; KMS / queue read-exposure reaches GDPR **Art.32(1)(b)**). The skill's version blurb + the `compliance_check` capability row are refreshed to seven frameworks (corrected a stale "six/hexa-framework" reference); plugin count UNCHANGED at 28; all seven coverage matrices UNCHANGED. Paired with EE 0.30.0 + CE 0.2.12.
|
|
211
|
+
Paired content bump — no new framework. EE 0.30.0 is a detection-depth + mapping-correctness release: it closes a wave of cloud-misconfiguration false-negatives across AWS (S3 access-points · resource-policy effective-exposure · EC2-SG public-CIDR/split-range · KMS effective-decrypt/cross-account · IAM ListUsers truncation) and Azure (storage/NSG/Key-Vault/cloud-scanner class-O fail-opens), and routes architecturally-identical confidentiality / least-privilege exposures consistently across all seven frameworks (e.g. a public application entry-point now appears on a NIST CSF **PR.AA-05** least-privilege report; KMS / queue read-exposure reaches GDPR **Art.32(1)(b)**). The skill's version blurb + the `compliance_check` capability row are refreshed to seven frameworks (corrected a stale "six/hexa-framework" reference); plugin count UNCHANGED at 28; all seven coverage matrices UNCHANGED. Paired with EE 0.30.0 + CE 0.2.12. *(WITHDRAWN at 0.2.34 — the tool named here does not exist; this line is the historical record of what the package said at the time.)*
|
|
171
212
|
|
|
172
213
|
## 0.2.11 (2026-06-12) — GDPR Article 32 framework taught (paired with CE 0.2.11 + EE 0.20.0)
|
|
173
214
|
|
|
@@ -299,7 +340,7 @@ Paired-release pin for the EE 0.13.1 cycle: CIS-Hardened-Image detection goes LI
|
|
|
299
340
|
|
|
300
341
|
## 0.1.42 (PUBLISHED 2026-05-24) — Paired-release pin for EE 0.13.0 + CE 0.1.75 — CIS Critical Security Controls v8 sixth-framework introduction
|
|
301
342
|
|
|
302
|
-
**Cycle hook**: EE 0.13.0 ships CIS Critical Security Controls v8 (Center for Internet Security, May 2021; v8.1 errata June 2024) as the sixth Track 3 framework — **17 covered + 21 partial + 115 OOS across 153 Safeguards / 18 Controls / 3 cumulative Implementation Groups** (engine substrate IG1 23-of-56 / IG2-cumulative 36-of-130 / IG3-cumulative 38-of-153). Implementation Group cumulative discipline (IG1=56 cyber-insurance baseline / IG2 cumulative=130 / IG3 cumulative=153; smallest-IG-membership tagging) + no-certification-body attestation discipline (CSAT / CIS-CAT Pro self-attestation, never "CIS certified") + Cloud Companion Guide v8 shared-responsibility + CIS-Hardened-Image substrate-evidence credit (4.1/4.2/4.6) + 5 Security Functions NOT 6 (no Govern) + 6 Asset Types + MS-ISAC/EI-ISAC/H-ISAC sector baselines + v7.1-to-v8 cross-reference. a dedicated framework audit review authored 2026-05-24 via /skill-creator (833 lines / 5 files) per the institutional Per-Framework Adversarial-Audit Skill Pairing pattern. **`compliance_check` SKILL.md row updated FIVE → SIX shipped frameworks** with CIS Safeguard examples + IG-cumulative + no-cert-body attestation framing. No other agent-skill code changes — paired-publish for trio-publish discipline + customer discoverability.
|
|
343
|
+
**Cycle hook**: EE 0.13.0 ships CIS Critical Security Controls v8 (Center for Internet Security, May 2021; v8.1 errata June 2024) as the sixth Track 3 framework — **17 covered + 21 partial + 115 OOS across 153 Safeguards / 18 Controls / 3 cumulative Implementation Groups** (engine substrate IG1 23-of-56 / IG2-cumulative 36-of-130 / IG3-cumulative 38-of-153). Implementation Group cumulative discipline (IG1=56 cyber-insurance baseline / IG2 cumulative=130 / IG3 cumulative=153; smallest-IG-membership tagging) + no-certification-body attestation discipline (CSAT / CIS-CAT Pro self-attestation, never "CIS certified") + Cloud Companion Guide v8 shared-responsibility + CIS-Hardened-Image substrate-evidence credit (4.1/4.2/4.6) + 5 Security Functions NOT 6 (no Govern) + 6 Asset Types + MS-ISAC/EI-ISAC/H-ISAC sector baselines + v7.1-to-v8 cross-reference. a dedicated framework audit review authored 2026-05-24 via /skill-creator (833 lines / 5 files) per the institutional Per-Framework Adversarial-Audit Skill Pairing pattern. **`compliance_check` SKILL.md row updated FIVE → SIX shipped frameworks** with CIS Safeguard examples + IG-cumulative + no-cert-body attestation framing. No other agent-skill code changes — paired-publish for trio-publish discipline + customer discoverability. *(WITHDRAWN at 0.2.34 — the tool named here does not exist; this line is the historical record of what the package said at the time.)*
|
|
303
344
|
|
|
304
345
|
**Plugin catalog**: UNCHANGED at 24 plugins; MCP tool signatures unchanged; schemas unchanged; workflows unchanged. **SOC 2 + HIPAA + NIST CSF + PCI DSS + ISO 27001 matrices ALL UNCHANGED**; **CIS Controls v8 matrix NEW at 17/21/115 across 153 Safeguards**.
|
|
305
346
|
|
|
@@ -331,7 +372,7 @@ Paired-release pin for the EE 0.13.1 cycle: CIS-Hardened-Image detection goes LI
|
|
|
331
372
|
|
|
332
373
|
**Cycle hook**: EE 0.11.0 introduces PCI DSS v4.0.1 (PCI SSC, June 2024 errata; supersedes v4.0 March 2022; v3.2.1 retired March 31, 2024) as the fourth compliance framework alongside SOC 2 (AICPA TSC 2017), HIPAA Security Rule §164.312, and NIST Cybersecurity Framework 2.0. The agent-skill catalog updates accordingly:
|
|
333
374
|
|
|
334
|
-
- `compliance_check` MCP tool description widened from "SOC 2 + HIPAA + NIST CSF 2.0" to "SOC 2 + HIPAA + NIST CSF 2.0 + PCI DSS v4.0.1" with the matching `--compliance soc2,hipaa,nist-csf,pci-dss` CSV invocation hint. PCI DSS sub-requirement examples baked into tool description: `Req 1.2.1` NSC config standards, `Req 8.4.1` MFA on non-console admin, `Req 10.2.1` audit logs enabled, `Req 11.3.1` quarterly internal vuln scans. Defined-vs-Customized Approach discipline per PCI DSS v4.0.1 Appendix E (15 Defined-only sub-requirements enforced at schema layer) + CHD Scope operator-attested via CDE Data Flow Diagram per Req 1.2.4 + card-brand AOC enforcement view (Visa CISP / Mastercard SDP / Amex DSOP / Discover DISC).
|
|
375
|
+
- `compliance_check` MCP tool description widened from "SOC 2 + HIPAA + NIST CSF 2.0" to "SOC 2 + HIPAA + NIST CSF 2.0 + PCI DSS v4.0.1" with the matching `--compliance soc2,hipaa,nist-csf,pci-dss` CSV invocation hint. PCI DSS sub-requirement examples baked into tool description: `Req 1.2.1` NSC config standards, `Req 8.4.1` MFA on non-console admin, `Req 10.2.1` audit logs enabled, `Req 11.3.1` quarterly internal vuln scans. Defined-vs-Customized Approach discipline per PCI DSS v4.0.1 Appendix E (15 Defined-only sub-requirements enforced at schema layer) + CHD Scope operator-attested via CDE Data Flow Diagram per Req 1.2.4 + card-brand AOC enforcement view (Visa CISP / Mastercard SDP / Amex DSOP / Discover DISC). *(WITHDRAWN at 0.2.34 — the tool named here does not exist; this line is the historical record of what the package said at the time.)*
|
|
335
376
|
- `SKILL.md` framework-coverage table extended with PCI DSS v4.0.1 sub-requirement-level matrix (**20 covered / 8 partial / 39 OOS across 67 of ~250 sub-requirements at MVP-67 density**).
|
|
336
377
|
- `references/plugins.md` framework-bullet extended from "three compliance frameworks" to "four compliance frameworks" with PCI DSS v4.0.1 sub-requirement examples + Req 12 OOS-by-design entirely framing + Req 5 + Req 9 OOS-entirely framing + Drata PCI / Vanta PCI / AuditBoard PCI / OneTrust GRC pairing-platform names.
|
|
337
378
|
|
|
@@ -343,7 +384,7 @@ Paired-release pin for the EE 0.13.1 cycle: CIS-Hardened-Image detection goes LI
|
|
|
343
384
|
|
|
344
385
|
**Cycle hook**: EE 0.10.0 introduces NIST Cybersecurity Framework 2.0 (NIST CSWP 29, February 2024) as the third compliance framework alongside SOC 2 (AICPA TSC 2017) and HIPAA Security Rule §164.312. The agent-skill catalog updates accordingly:
|
|
345
386
|
|
|
346
|
-
- `compliance_check` MCP tool description widened from "SOC 2 + HIPAA" to "SOC 2 + HIPAA + NIST CSF 2.0" with the matching `--compliance soc2,hipaa,nist-csf` CSV invocation hint.
|
|
387
|
+
- `compliance_check` MCP tool description widened from "SOC 2 + HIPAA" to "SOC 2 + HIPAA + NIST CSF 2.0" with the matching `--compliance soc2,hipaa,nist-csf` CSV invocation hint. *(WITHDRAWN at 0.2.34 — the tool named here does not exist; this line is the historical record of what the package said at the time.)*
|
|
347
388
|
- `SKILL.md` framework-coverage table extended with NIST CSF 2.0 Subcategory-level matrix (13 covered / 10 partial / 83 OOS across 106 of CSF 2.0's 107 Subcategories).
|
|
348
389
|
- `references/plugins.md` framework-bullet extended from "two compliance frameworks" to "three compliance frameworks" with NIST CSF 2.0 control-ID examples (PR.AA-01, DE.CM-01, RC.RP-03) + Implementation Tiers OOS disclaimer explanation + Tugboat Logic / Drata / Vanta / AuditBoard pairing-platform names.
|
|
349
390
|
|
package/README.md
CHANGED
|
@@ -115,7 +115,7 @@ This package provides **knowledge about** NSAuditor AI. To actually **run** scan
|
|
|
115
115
|
|---------|-------|-----------|
|
|
116
116
|
| **Community** | Free / MIT | 27 plugins (service probes + host/network discovery + intelligence/meta), basic AI, SARIF, CTEM, scan history |
|
|
117
117
|
| **Pro** | $49/mo | + CVE matching, risk scoring, 3 Pro plugins (040 TLS / 050 TRIBE / 060 DNS) |
|
|
118
|
-
| **Enterprise** | $2k+/yr | + 28
|
|
118
|
+
| **Enterprise** | $2k+/yr | + 28 enterprise plugins (1020-1222 range) — 27 cloud-substrate auditors across AWS / Azure / GCP plus `1023 Zero Trust Assessment`, which scores zero-trust posture from a NETWORK-host scan and does not run on a cloud pass, seven-framework evidence-pack (SOC 2 / HIPAA / NIST CSF 2.0 / PCI DSS v4.0.1 / ISO 27001:2022 / CIS Controls v8 / GDPR Art. 32), SHA-256 chain-of-custody attestations, air-gapped operation (offline licensing + offline CVE matching) |
|
|
119
119
|
|
|
120
120
|
→ [Pricing](https://www.nsauditor.com/ai/pricing/)
|
|
121
121
|
|
package/SKILL.md
CHANGED
|
@@ -16,7 +16,9 @@ description: >
|
|
|
16
16
|
|
|
17
17
|
# NSAuditor AI — Agent Skill
|
|
18
18
|
|
|
19
|
-
> **Version:** 0.2.
|
|
19
|
+
> **Version:** 0.2.36 (post-EE-0.33.0 · **requires CE ≥ 0.2.37**) — a wording-only patch on top of the wiring release: the RFC 3161 row said the capability was "implemented but not yet wired to a flag", which EE 0.33.0 falsified and the live-TSA smokes then falsified twice over (npm path 2026-08-07, from inside the `:0.33.0` container image 2026-08-08). It is opt-in via `NSAUDITOR_TSA_URL`, proven on both delivery vehicles, and the only surviving caveat is version scope — retained images `:0.32.11` and earlier carry no `openssl`. **Prior:** 0.2.35 — the wiring release. Three CE entry points are new and the skill should teach them: `compliance attest` (Type II recurring-scan attestation over a directory of prior scans; exit 3 on an empty history), and `--sla-policy <file>` / `--compliance-history <dir>`, which reach the SLA/MTTR engine. A startup posture veto refuses `NSAUDITOR_OFFLINE_ONLY=1` together with a configured outbound path (exit 2). ⚠️ Two corrections to this package's own prose: the `compliance_check` disclosure now carries the WITHDRAWN marker so an instrument can tell it from a claim, and the Enterprise row no longer calls plugin 1023 a cloud auditor — it scores zero-trust posture from a NETWORK-host scan and cannot be run by selecting it with `--plugins`. Plugin catalog UNCHANGED at 28 EE / 55 overall; all seven matrices UNCHANGED.
|
|
20
|
+
|
|
21
|
+
> **Prior:** 0.2.34 (post-EE-0.32.11 — **a correction release.** The Verification Engine was WITHDRAWN at EE 0.32.7; this package went on presenting it as a shipped Phase 4, glossing `VERIFIED` as probe-confirmed, and selling "verification probes" in the pricing table. All corrected — the status enum stays, the active-probe gloss goes, and findings are stated to be emitted UNVERIFIED. Upstream: `scan_cloud` summaries now surface the INFO tier and `deferredScope` boundaries; a new `compliance_matrix` MCP tool returns the shipped coverage matrix for any of the seven frameworks, derived at call time; `pdfExport` withdrawn and every capability now ships a description. Plugin catalog UNCHANGED at 28 EE / 55 overall; all seven matrices UNCHANGED.)
|
|
20
22
|
|
|
21
23
|
> **Prior:** 0.2.33 (post-EE-0.32.10 — no knowledge change.)
|
|
22
24
|
|
|
@@ -169,7 +171,7 @@ These tools return a license upgrade prompt on CE installations:
|
|
|
169
171
|
> **Reporting `scan_cloud` region scope — never overstate coverage:** Report the regions you ACTUALLY scanned, derived from the `regions` you **passed** — NOT from the findings. If you OMITTED `regions`, only the single server-default region (`AWS_REGION`) was scanned — say exactly that and add that the account's OTHER enabled regions were NOT covered (offer to re-run for all regions). **Never escalate a single-region or "quick" request into a multi-region scan.** Do NOT claim "all regions" / "every region" / "across N regions" just because GuardDuty or Inspector list per-region findings: those plugins enumerate every enabled region INTERNALLY regardless of scope, so their per-region findings are NOT evidence the other plugins ran outside the region(s) you passed.
|
|
170
172
|
|
|
171
173
|
> **Full all-region coverage — discover then batch** (use ONLY when the user explicitly asked for all/every/whole-account/complete/full region coverage; NEVER for a plain or "quick" request — those stay single-region per the `regions` default above): a single `regions:["all"]` call usually exceeds the host's MCP tool-call timeout (e.g. Claude Desktop's) and returns nothing. Reliable pattern: (1) run a default scan (omit `regions`) — its GuardDuty/Inspector findings enumerate every enabled region, giving you the full list while auditing the default region; (2) scan the REMAINING regions in small batches (3–5 region codes per `regions:[...]` call) across successive calls until every enabled region is covered; (3) merge and report the TOTAL number of regions actually covered — **count** them, don't guess. If you try `["all"]` and it times out, that result is INCOMPLETE — fall back to batching and continue until complete; never report a timed-out or partial scan as full coverage.
|
|
172
|
-
> **⚠️ COMPLIANCE IS A CLI SURFACE, NOT AN MCP TOOL.**
|
|
174
|
+
> **⚠️ COMPLIANCE IS A CLI SURFACE, NOT AN MCP TOOL.** `compliance_check` is WITHDRAWN — there is no such tool and there never was, and the evidence pack is produced by `nsauditor-ai scan --host <target> --compliance <fw> --out <dir>`, which runs the compliance PHASE and writes `scan_compliance_<framework>.{json,md,html}` plus its attestation and chain-of-custody sidecars. `scan_cloud` maps findings to frameworks but **never runs that phase**, so no MCP call produces a pack. Use `compliance_matrix` to state COVERAGE (what is covered / partial / out of scope); use the CLI to produce EVIDENCE.
|
|
173
175
|
>
|
|
174
176
|
> The framework detail below is accurate and worth keeping — it was attached to a tool name that does not exist:
|
|
175
177
|
>
|
|
@@ -220,6 +222,16 @@ mapping the engine did not emit, and never claim an MCP call produced a pack.
|
|
|
220
222
|
|
|
221
223
|
---
|
|
222
224
|
|
|
225
|
+
### Suppressions — the workflow SHIPS, the signature does NOT (post EE 0.33.1)
|
|
226
|
+
|
|
227
|
+
An operator can mark a finding as accepted-risk or false-positive. That suppression workflow **ships and is reachable**: a suppressed finding renders with status `FALSE_POSITIVE`, set by an operator rather than by a probe, and the suppression is applied before the pack is written — so it sits **inside** the hashed artifact rather than being independently attested.
|
|
228
|
+
|
|
229
|
+
⚠️ **Ed25519 SIGNING of suppressions is implemented in the package but NOT REACHABLE — no shipped entry point signs a suppression, so no first-party run produces a signed one.** The signer backends and the frozen `algorithm` / `backend` record fields are groundwork, deliberately landed before reachability because retrofitting algorithm agility once signatures exist in customer archives would break every auditor holding one. Setting `NSAUDITOR_SIGNING_KEY` changes nothing today, because nothing calls the signer.
|
|
230
|
+
|
|
231
|
+
**Never tell an operator their suppressions are cryptographically signed** — not even in the softer form "signed once you configure a key". If asked what the SHA-256 chain-of-custody covers: it proves the pack was not altered after generation. It does **not** attest who suppressed a finding, when, or with what justification, and an assessor asking about suppressions usually wants the second thing.
|
|
232
|
+
|
|
233
|
+
⚠️ **Disambiguation:** "suppression" also appears in this package in the unrelated **AWS SES email suppression list** sense (plugin 1190). They are different subjects; check which one is being asked about.
|
|
234
|
+
|
|
223
235
|
## Five-Phase Pipeline Architecture
|
|
224
236
|
|
|
225
237
|
NSAuditor AI follows an institutional five-phase pipeline:
|
|
@@ -430,7 +442,7 @@ Add to your MCP configuration with the same command/args pattern.
|
|
|
430
442
|
|---------|-------|-------------|
|
|
431
443
|
| **Community** | Free / MIT | 27 plugins (service probes + host/network discovery + intelligence/meta), basic AI, CTEM, SARIF, scan history |
|
|
432
444
|
| **Pro** | $49/mo | + CVE matching, risk scoring, analysis agents, Pro plugins (040 TLS / 050 TRIBE / 060 DNS) |
|
|
433
|
-
| **Enterprise** | $2k+/yr | + 28
|
|
445
|
+
| **Enterprise** | $2k+/yr | + 28 enterprise plugins (1020-1222 range) — 27 cloud-substrate auditors covering AWS / GCP / Azure plus `1023 Zero Trust Assessment` (which declares no cloud provider and scores zero-trust posture from a NETWORK-host scan — it never runs on a cloud pass, and selecting it by id on its own does not run it either) — against seven frameworks (SOC 2 10 covered + 4 partial; HIPAA; NIST CSF 2.0; PCI DSS v4.0.1; ISO/IEC 27001:2022; CIS Controls v8; GDPR Art. 32 infrastructure substrate); SOC 2 evidence-pack generation; SHA-256 chain-of-custody attestations (RFC 3161 timestamping is opt-in via the `NSAUDITOR_TSA_URL` environment variable — there is no CLI flag and no default, it makes an outbound call to the Time-Stamp Authority you name, and it was exercised against a live Time-Stamp Authority on BOTH delivery vehicles — the npm path and, from inside the pushed `:0.33.0` Marketplace image, on 2026-08-08; retained images `:0.32.11` and earlier carry no `openssl`); air-gapped operation (offline licensing + offline CVE matching under `NSAUDITOR_OFFLINE_ONLY=1`) |
|
|
434
446
|
|
|
435
447
|
→ [Pricing](https://www.nsauditor.com/ai/pricing/)
|
|
436
448
|
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "nsauditor-ai-agent-skill",
|
|
3
|
-
"version": "0.2.
|
|
3
|
+
"version": "0.2.36",
|
|
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",
|