New in EE 2.1.1 — GCP 1024 and 1025 down with a cause, openai 7 and one zod, a partial AI read stated, tier denials that name what you can use; paired with Community 1.1.1 →

Security Intelligence
Without Data Exposure.

Audit-ready compliance evidence from one read-only scan — SOC 2, HIPAA, NIST CSF 2.0, PCI DSS v4.0.1, ISO/IEC 27001:2022, CIS Controls v8, GDPR Article 32, and NIST SP 800-171 Rev 2, mapped from a single pass across AWS, Azure and GCP. 56 plugins, built on an open-source core. Runs entirely on your infrastructure — zero data exfiltration by architecture. Enterprise 2.1.1: the GCP Cloud Storage and IAM auditors report themselves down, with the cause, when no project is set or no usable credential loads; Community 1.1.1 moves to openai 7 with one zod and says when an AI assessment covers only part of the scan.

nsauditor-ai — scan
$ nsauditor-ai scan --host aws --plugins all --compliance soc2,hipaa,nist-csf,pci-dss,iso-27001,cis-v8,gdpr,nist-800-171
✓ 56 plugins loaded (27 CE + 29 EE)
✓ Frameworks: SOC 2 · HIPAA §164.312 · NIST CSF 2.0 · PCI DSS v4.0.1 · ISO 27001:2022 · CIS Controls v8 · GDPR Art. 32 · NIST SP 800-171
✓ scan_compliance_soc2.{md,html,json} → 10/4/37
✓ scan_compliance_hipaa.{md,html,json} → 7/3/45
✓ scan_compliance_nist-csf.{md,html,json} → 13/10/83
✓ scan_compliance_pci-dss.{md,html,json} → 19/9/44 · 72 enumerated
✓ scan_compliance_iso-27001.{md,html,json} → 17/14/62
✓ scan_compliance_cis-v8.{md,html,json} → 17/23/113
✓ scan_compliance_gdpr.{md,html,json} → 4/5/2 Art.32 substrate
✓ Any cloud that could not be scanned is recorded as a fail-closed evidence gap — never as a pass
✓ Zero data exfiltration — no findings left your infra
# Audit a cloud account directly from Claude Desktop (MCP) — "audit my AWS account"
$ nsauditor-ai scan --host aws --env ~/envs/prod.env --compliance soc2
$ nsauditor-ai scan --host aws --aws-profile prod --compliance soc2
Get Started

The cloud-audit + compliance engine shown above is NSAuditor AI Enterprise · Community Edition is free forever, MIT licensed

What's Inside
🔒

Zero Data Exfiltration

Runs entirely on your machine. No cloud. No telemetry. License validation is offline. We can't see your data because we never touch it.

🎯

Risk-Scored Prioritization

Findings are risk-scored and ranked so you fix what matters first. Suppress accepted-risk or false-positive findings with the operator workflow — your triage decisions persist across scans.

🤖

AI-Powered Analysis

OpenAI, Claude, or Ollama (fully local). Compliance reports, remediation guidance, risk prioritization. Your API keys, your data.

📊

56 Scanner Plugins

27 CE plugins: Ports, SSH, HTTP, TLS, DNS, SNMP, SMB, RPC, mDNS, UPnP, and more. 29 EE plugins, of which 28 are cloud auditors: AWS S3, GCP, Azure, IAM Deep Auditor, CloudTrail, API Gateway, DynamoDB, KMS, Lambda, Secrets+SSM, CodePipeline, IAM Decrypt-Path, S3 Lifecycle, AWS Backup, RDS, SES, VPC/PrivateLink, EC2 SG, ElastiCache, Inspector2/GuardDuty, plus dedicated Azure Storage / NSG perimeter / Key Vault deep auditors, and more. The 29th, Zero Trust Assessment, is not a cloud auditor — it scores zero-trust posture from a network-host scan and calls no cloud API.

📋

Octa-Framework Compliance

SOC 2 (AICPA TSC 2017), HIPAA Security Rule §164.312, NIST CSF 2.0, PCI DSS v4.0.1, ISO/IEC 27001:2022, CIS Critical Security Controls v8, GDPR Article 32 (Security of Processing), and NIST SP 800-171 Rev 2 — all from one scan. Sub-requirement-level mapping for QSA RoC workflow. Customized Approach eligibility read from each requirement's own objective cell, identifiers derived from the standard. CHD Scope operator-attested. ISO 27001 Statement of Applicability discipline. CIS Implementation Group cumulative discipline (IG1 cyber-insurance baseline). GDPR Article 32 is an infrastructure substrate for Art. 32 only (4 covered + 5 partial + 2 OOS across 11 sub-measure units) — NOT GDPR compliance. Auditor-ready evidence packs with SHA-256 chain-of-custody. No BAA with Nsasoft required.

🔌

MCP Integration — free in Community

The MCP server ships free in the Community Edition — drive NSAuditor from Claude Desktop, Claude Code, Cursor, or any MCP-aware agent. Run nsauditor-ai mcp install-key and paste the mcpServers block it prints into your claude_desktop_config.json (it names node and the server script by absolute path), or in Claude Code: claude mcp add nsauditor-ai --env NSA_MCP_AUTH_KEY=<from: nsauditor-ai mcp install-key> -- nsauditor-ai-mcp. Then install the optional agent skill so the assistant knows NSAuditor's tools, schemas, and audit workflows — in Claude Desktop: Skills → Create skill → Upload a skill (upload SKILL.md). Then just ask: "audit my AWS account."

Latest Release — EE 2.1.1 · CE 1.1.1 · agent-skill 1.1.2 · 2026-11-01
📖

EE 2.1.1 — GCP 1024 and 1025 down with a cause, openai 7 and one zod, a partial AI read stated, tier denials that name what you can use; paired with Community 1.1.1

The GCP Cloud Storage (1024) and IAM (1025) auditors report themselves down, with the cause, when no GCP project is set or no usable credential loads, as the GCP security audit (1021) does. Community 1.1.1 moves its OpenAI client to openai 7, so a project that depends on it holds one zod at 4.6.6 or later, with no overrides; its AI stage says at the head of the assessment when the provider reports reading less than half of the input, by the stage’s own estimate; a tier denial names the tools the caller can use now; and nsauditor-ai license with an unknown subcommand, or none, exits 2. Enterprise 2.1.1 is a patch release: no plugin added or removed, and no coverage-matrix, peer-floor or Node.js runtime change. See Enterprise →

GCP plugins that report themselves down (Enterprise). With no GCP project set, or no usable credential, 1024 and 1025 return up: false with the cause 1021 gives, and call no API. In 2.1.0 they reported themselves up, and with credentials but no project named, 1024 could audit whatever project Google’s client library found on its own. The controls they map fail closed as not measured. openai 7 and one zod (Community 1.1.1). A project that depends on Community holds one zod at 4.6.6 or later, with no overrides, and no node-domexception from Community. The AI stage was run end to end on openai 7 against OpenAI and a local Ollama model before the release. A partial AI read is stated (Community 1.1.1). When the provider reports reading less than half of the input, by the CLI’s own estimate, the assessment files open with one line saying the assessment covers part of the scan, and the JSON response and the run record carry both numbers. A local Ollama model with default settings read about a tenth of a host scan. Tier denials name what you can use (Community 1.1.1). compliance_matrix without the Enterprise package names the licence it needs instead of an install command that fails, and nsauditor-ai license with an unknown subcommand, or none, exits 2. The 2.1.1 air-gap carrier ships Enterprise 2.1.1 with Community 1.1.1; replace a 2.1.0 carrier with it. The agent skill 1.1.2 teaches the release.

Where it stops, stated: without impersonation, resolving a GCP credential fetches no token: a key file that loads but is revoked, or a principal without the IAM permission, still fails at the first call and stays a LOW evidence gap with up: true. The partial-read check estimates input as characters ÷ 4 for every provider, and the threshold was not measured on the Claude provider. With Community 1.1.0, which Enterprise 2.1.1 still accepts, a project root still holds zod 3.25.76: install Community 1.1.1 with it. The two measured limits stated since Enterprise 1.3.0 still hold: a scan made before 1.3.0 could not record a plugin left out of it, and a scan that found its ports with the Nmap plugin (024) alone has no port oracle, so include the port scanner (003) in the scans you compare.

Upgrade both packages together: Enterprise 2.1.1 runs on Node.js 24.22.1 or newer and still requires Community 1.0.0 or newer; install Community 1.1.1 with it — npm i -g nsauditor-ai@1.1.1 @nsasoft/nsauditor-ai-ee@2.1.1. Install Node.js 24.22.1 or newer first: on an older runtime npm prints an EBADENGINE warning and installs anyway, and the air-gap install.sh refuses. Enterprise 2.1.1, like 2.1.0 and 2.0.0, imports a name that first ships in Community 1.0.0; below it Enterprise’s core does not load while its plugins still do — on Community 0.2.57 as on 0.2.56 — so the scan skips its intelligence, analysis-agent and compliance stages. Do not name a Community older than 1.0.0 in the same npm install -g command as Enterprise: npm then nests a second Community under Enterprise, and Enterprise’s stages are skipped with no message and no compliance report (measured on 2.1.0). Install the paired versions together, or Community first. An estate scanned with 2.1.0 and then with 2.1.1 with no GCP project or no usable credential: the earlier scan’s 1024 and 1025 violations are withheld from MTTR closure, and the drift file lists them as could not measure, never as cleared. Coverage: all eight coverage matrices unchanged since EE 2.1.0. 29 Enterprise auditors (28 cloud auditors + 1 non-cloud), 56 plugins overall; floor CE ≥ 1.0.0. Release notes live in the package CHANGELOG.md.

Viewing Scan Reports
out/<host>_<timestamp>/ — output files
scan_response_ai.html Styled AI report — CVE links, risk badges, remediation guidance
scan_conclusion_raw.html Full admin detail — interactive filters, all plugin findings unredacted
scan_intelligence.json CVE matches, MITRE ATT&CK mapping, risk scores (Pro)
scan_conclusion_raw.json Full machine-readable conclusion for automation
scan_response_ai.txt AI analysis as plain Markdown — paste into issues, Slack, chat
scan_compliance_soc2.{html,json,md} SOC 2 evidence pack — AICPA TSC 2017, SHA-256 chain-of-custody (EE)
scan_compliance_hipaa.{html,json,md} HIPAA §164.312 evidence pack — R/A discipline, no BAA with Nsasoft (EE)
scan_compliance_nist-csf.{html,json,md} NIST CSF 2.0 evidence pack — subcategory-level, SP 800-53 refs (EE)
scan_compliance_pci-dss.{html,json,md} PCI DSS v4.0.1 evidence pack — sub-requirement-level for QSA RoC, CHD Scope operator-attested, card-brand AOC enforcement view (EE)
scan_compliance_gdpr.{html,json,md} GDPR Article 32 evidence pack — Security-of-Processing infrastructure substrate (Art. 32 only, not GDPR compliance), 4 covered + 5 partial + 2 OOS across 11 sub-measure units (EE)
scan_drift_<fw>.json What changed since the previous compliance scan — controls to fail or to pass, controls that lost measurement, violations new or cleared, and what could not be measured, with its reason (EE)
Open reports in your browser
# macOS — open AI report open out/192.168.1.1_*/scan_response_ai.html # macOS — open full admin detail open out/192.168.1.1_*/scan_conclusion_raw.html # Linux xdg-open out/192.168.1.1_*/scan_response_ai.html # Custom output directory nsauditor-ai scan --host 10.0.0.1 --out ./reports open ./reports/10.0.0.1_*/scan_response_ai.html # Markdown report (paste-ready) nsauditor-ai scan --host 10.0.0.1 --output-format md
56
Scanner Plugins
8
Compliance Frameworks
MIT
Licensed
0
Data Collected
Three Editions
Community
Free
forever, MIT licensed
npm install -g nsauditor-ai
  • 27 scanner plugins
  • AI analysis (OpenAI, Claude, Ollama)
  • CTEM watch mode
  • SARIF + CSV export
  • MCP server (Claude Desktop / Code / Cursor)
Pro
$39/mo
billed $470/yr · save 20% vs monthly
Learn More & Subscribe →
  • CVE matching + MITRE
  • Parallel analysis agents
  • Finding queue + suppression workflow
  • Risk scoring
  • Exploit intelligence — CISA KEV + FIRST EPSS (stores you populate)
  • Send-ready reports — report --from <dir> --format executive → self-contained executive HTML, or --format jira → a Jira-importer CSV
  • Delta reports — what changed since the last scan, and a not-comparable list with the reason on every row
  • Intelligence-enriched AI reports