blackveil-dns 3.100.1 → 3.117.3

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/README.md CHANGED
@@ -7,7 +7,7 @@
7
7
  Source-available DNS & email security scanner for Claude, Cursor, VS Code, and MCP clients across Streamable HTTP, stdio, and legacy HTTP+SSE.
8
8
 
9
9
  [![GitHub stars](https://img.shields.io/github/stars/MadaBurns/bv-mcp?style=flat&logo=github)](https://github.com/MadaBurns/bv-mcp/stargazers)
10
- [![MCP tools](https://img.shields.io/badge/MCP%20tools-81-brightgreen)](https://github.com/MadaBurns/bv-mcp/actions)
10
+ [![MCP tools](https://img.shields.io/badge/MCP%20tools-82-brightgreen)](https://github.com/MadaBurns/bv-mcp/actions)
11
11
  [![BUSL-1.1 License](https://img.shields.io/badge/License-BUSL--1.1-blue.svg)](LICENSE)
12
12
  [![MCP](https://img.shields.io/badge/MCP-2025--06--18-blue)](https://modelcontextprotocol.io/)
13
13
  [![Cloudflare Workers](https://img.shields.io/badge/Cloudflare%20Workers-F38020?logo=cloudflare&logoColor=white)](https://workers.cloudflare.com/)
@@ -25,7 +25,7 @@ Source-available DNS & email security scanner for Claude, Cursor, VS Code, and M
25
25
 
26
26
  **Claude Desktop** (one-click install):
27
27
 
28
- Download the [Blackveil DNS extension](https://github.com/MadaBurns/bv-claude-dns/releases/latest/download/bv-claude-dns.mcpb) and open it — the current 81-tool surface is available instantly. [Verify your download](https://blackveilsecurity.com/extensions/claude-dns#install).
28
+ Download the [Blackveil DNS extension](https://github.com/MadaBurns/bv-claude-dns/releases/latest/download/bv-claude-dns.mcpb) and open it — the current 82-tool surface is available instantly. [Verify your download](https://blackveilsecurity.com/extensions/claude-dns#install).
29
29
 
30
30
  **Claude Code** (one command):
31
31
 
@@ -67,7 +67,7 @@ For Streamable HTTP, clients should retain the `Mcp-Session-Id` returned by `ini
67
67
 
68
68
  ## What you get
69
69
 
70
- - **81 MCP tools with 19 scoring categories** — SPF, DMARC, DKIM, DNSSEC, SSL/TLS, MTA-STS, NS, CAA, MX, BIMI, TLS-RPT, subdomain takeover, HTTP security headers, DANE, SVCB/HTTPS, DANE-HTTPS, subdomailing, reverse DNS (PTR/FCrDNS), and DNSKEY strength
70
+ - **82 MCP tools with 19 scoring categories** — SPF, DMARC, DKIM, DNSSEC, SSL/TLS, MTA-STS, NS, CAA, MX, BIMI, TLS-RPT, subdomain takeover, HTTP security headers, DANE, SVCB/HTTPS, DANE-HTTPS, subdomailing, reverse DNS (PTR/FCrDNS), and DNSKEY strength
71
71
  - **Maturity staging** — Stage 0-4 classification (Unprotected to Hardened) with score-based capping to prevent inflated labels
72
72
  - **Trust surface analysis** — detects shared SaaS senders in SPF, both cataloged platforms (Google, M365, SendGrid) and uncataloged hosts identifiable by their `spf` delegation label, then cross-references DMARC enforcement to determine real exposure
73
73
  - **Guided remediation** — `generate` (artifact=`fix_plan`) produces provider-aware prioritized actions; its record artifacts (`spf_record`, `dmarc_record`, `dkim_config`, `mta_sts_policy`, `rollout_plan`) output ready-to-publish records; `validate_fix` confirms whether a fix was applied successfully
@@ -83,7 +83,7 @@ For Streamable HTTP, clients should retain the `Mcp-Session-Id` returned by `ini
83
83
  ## Tools
84
84
 
85
85
  ```
86
- 81 MCP tools · 7 prompts · 6 resources
86
+ 82 MCP tools · 7 prompts · 6 resources
87
87
 
88
88
  Email Auth Infrastructure Brand & Threats Meta
89
89
  ───────────── ────────────── ─────────────── ───────────────
@@ -281,11 +281,14 @@ The free tier requires no authentication. Authenticated requests bypass per-IP r
281
281
  - **Header**: `Authorization: Bearer <KEY>`
282
282
  - **Header (alternative)**: `X-API-Key: <KEY>` — for clients that cannot set `Authorization`. If both are sent, `Authorization: Bearer` wins.
283
283
  - **OAuth 2.1**: optional authorization-code flow with PKCE, enabled only when operators set `ENABLE_OAUTH=true`; owner-key consent is separately gated by `ENABLE_OWNER_OAUTH=true`.
284
+ - **Enterprise Managed Auth**: when OAuth is enabled, Claude Team and Enterprise organizations can connect through their identity provider with no per-user sign-in, using the RFC 7523 `jwt-bearer` grant (an ID-JAG assertion) and Claude's client-ID-metadata-document (CIMD) identity. Each customer issuer is an explicit allowlisted tenant; tokens never carry the `owner` tier. Admin setup: [**docs/enterprise-managed-auth.md**](docs/enterprise-managed-auth.md). The silent exchange starts from the HTTP 401 challenge described under paid-only tools below.
284
285
 
285
286
  The `?api_key=<KEY>` fallback is legacy/self-host compatibility only. BlackVeil hosted production sets `REJECT_QUERY_API_KEY=true`; clients that cannot send headers should use OAuth or an `mcp-remote` header bridge.
286
287
 
287
288
  For full hosted setup examples, stdio usage, OAuth setup, and legacy fallback endpoints, see [**docs/client-setup.md**](docs/client-setup.md).
288
289
 
290
+ To build the service into a product or a scheduled pipeline (storing scores, partial scans, batch jobs, retries, authority to scan), see the [**integrator guide**](docs/integrator-guide.md).
291
+
289
292
  ---
290
293
 
291
294
  ## CLI
@@ -304,7 +307,7 @@ blackveil evidence verify baseline.json
304
307
 
305
308
  Commands: `scan`, `check`, `batch`, `policy`, `drift save|compare`, `evidence verify`. Output formats (`--format`): `human` (default), `json`, `ndjson`, `evidence`; `--out <file>` writes to a file, `--fail-below <0-100>` or `--policy <file>` turns a scan into a pass/fail gate. Exit codes: `0` pass, `1` verified policy or integrity failure, `2` usage/input, `3` auth/quota/transport/tool error, `4` ungraded/inconclusive. `blackveil --help` prints the full usage (source of truth: `src/cli/command.ts`).
306
309
 
307
- The npm package `blackveil-dns` is frozen at 2.13.0 (published 2026-05-12), and that artifact carries only the `blackveil-dns-mcp` stdio bin, not `blackveil`, so `npm i -g blackveil-dns` does not install this CLI. Publishing is closed by policy ([#719](https://github.com/MadaBurns/bv-mcp/issues/719), closed as not planned). Build from source (`npm ci && npm run build`) for the CLI, or use [blackveil-dns-action](https://github.com/MadaBurns/blackveil-dns-action) in CI; it calls the hosted endpoint and needs no npm package.
310
+ The npm package `blackveil-dns` is published from release tags by a manually dispatched workflow, and not every release is published, so npm can be behind this repository: `npm view blackveil-dns version` shows what is there. Versions from 3.100.1 carry both the `blackveil` CLI and the `blackveil-dns-mcp` stdio bin; 2.13.0 carries only `blackveil-dns-mcp`. Build from source (`npm ci && npm run build`) for the CLI at the current commit, or use [blackveil-dns-action](https://github.com/MadaBurns/blackveil-dns-action) in CI; it calls the hosted endpoint and needs no npm package.
308
311
 
309
312
  ---
310
313
 
@@ -349,15 +352,15 @@ These live under the internal auth gate (`/internal/*`) and are called by bv-web
349
352
  | | **Free** | **Developer** | **Enterprise** |
350
353
  | -------------- | ---------------------------------------- | ---------------------------------------------------------- | ------------------------------------------- |
351
354
  | **Price** | $0 | $39/mo (provisioned by BlackVeil, not self-serve checkout) | [Contact us](https://blackveilsecurity.com) |
352
- | **Scans/day** | 25 (anonymous, per IP); 50 with a free key | 500 per tool | 10,000+ per tool |
353
- | **Checks/day** | Tool-specific limits | Tool-specific limits | Contract limits |
355
+ | **Scans/day** | 25 (anonymous, per IP); 50 with a free key | 500 per tool, 5,000 across all tools | 10,000+ per tool, 50,000 across all tools |
356
+ | **Checks/day** | Tool-specific limits | Tool-specific limits | Tool-specific limits |
354
357
  | **Rate limit** | 50 req/min | None | None |
355
358
  | **API access** | No | Yes | Yes |
356
359
  | **MCP access** | Yes | Yes | Yes |
357
360
 
358
- Free-tier daily limits depend on whether you send a key. Anonymous (no key): 25 `scan_domain` scans/day per IP, across at most 12 distinct domains/day. Free key: 50/day per tool. Free keys do not carry the REST scan API (see [REST / HTTP](#rest--http)); "API access" above means that keyed API. Limits are read from `TIER_DAILY_LIMITS`, `FREE_TOOL_DAILY_LIMITS` and `FREE_DISTINCT_DOMAIN_DAILY_LIMIT` in `src/lib/config.ts`.
361
+ Free-tier daily limits depend on whether you send a key. Anonymous (no key): 25 `scan_domain` scans/day per IP, across at most 12 distinct domains/day. Free key: 50/day per tool. Agent, developer and enterprise keys also have a daily total across all tools combined (2,000, 5,000 and 50,000 calls respectively), on top of their per-tool limits; the first call past the total gets HTTP 429. Free keys do not carry the REST scan API (see [REST / HTTP](#rest--http)); "API access" above means that keyed API. Limits are read from `TIER_DAILY_LIMITS`, `TIER_AGGREGATE_DAILY_LIMITS`, `FREE_TOOL_DAILY_LIMITS` and `FREE_DISTINCT_DOMAIN_DAILY_LIMIT` in `src/lib/config.ts`.
359
362
 
360
- Offensive/recon and multi-domain tools (subdomain discovery, attack-path simulation, lookalike/shadow-domain detection, fast-flux detection, supply-chain mapping, real-time threat feed, bucket/OSINT investigations, `batch_scan`, `compare_domains`, brand audits) require a paid plan (Developer tier or higher); free, unauthenticated, and agent-tier callers get an HTTP 403 upgrade-required response. Unauthenticated callers are additionally capped at 12 distinct domains per day (best-effort, fail-open). The OSINT/bucket status and report pollers stay free; the brand-audit and brand-discovery pollers (`brand_audit_status`, `brand_audit_get_report`, `discover_brand_domains_status`, `discover_brand_domains_findings`) have no unauthenticated allowance and return HTTP 429 without a key.
363
+ Offensive/recon and multi-domain tools (subdomain discovery, attack-path simulation, lookalike/shadow-domain detection, fast-flux detection, supply-chain mapping, real-time threat feed, bucket/OSINT investigations, `batch_scan`, `compare_domains`, brand audits) require a paid plan (Developer tier or higher). An unauthenticated call gets HTTP 401 with a `WWW-Authenticate` challenge pointing at the OAuth protected-resource metadata (Claude shows its Connect card); a signed-in free or agent-tier caller gets an HTTP 403 upgrade-required response (`-32003`). Self-hosts without OAuth enabled keep the 403 for unauthenticated callers. Unauthenticated callers are additionally capped at 12 distinct domains per day (best-effort, fail-open). The OSINT/bucket status and report pollers stay free; the brand-audit and brand-discovery pollers (`brand_audit_status`, `brand_audit_get_report`, `discover_brand_domains_status`, `discover_brand_domains_findings`) have no unauthenticated allowance and return HTTP 429 without a key.
361
364
 
362
365
  ---
363
366
 
@@ -402,6 +405,6 @@ The supported self-host path is the stdio bin (`blackveil-dns-mcp`) for non-comm
402
405
 
403
406
  Built and maintained by [**BLACKVEIL**](https://blackveilsecurity.com) — NZ-owned cybersecurity consultancy.
404
407
 
405
- [Privacy Policy](https://www.blackveilsecurity.com/privacy) · [License](LICENSE) (BUSL-1.1 → MIT on 2030-03-17)
408
+ [Privacy Policy](https://www.blackveilsecurity.com/privacy) · [Terms of Service](https://www.blackveilsecurity.com/terms) · [License](LICENSE) (BUSL-1.1 → MIT on 2030-03-17)
406
409
 
407
410
  </div>
package/dist/cli.js CHANGED
@@ -103,6 +103,7 @@ var ToolListSchema = z.object({
103
103
  z.object({
104
104
  name: z.string().min(1),
105
105
  outputSchema: z.record(z.string(), z.unknown()).optional(),
106
+ inputSchema: z.object({ properties: z.record(z.string(), z.unknown()).optional() }).passthrough().optional(),
106
107
  annotations: z.object({ readOnlyHint: z.boolean().optional() }).passthrough().optional()
107
108
  }).passthrough()
108
109
  )
@@ -807,6 +808,12 @@ async function runScan(parsed, deps) {
807
808
  if (scanCode !== 0 || !baseline) return scanCode;
808
809
  return (await callPolicy(client, domain, baseline, deps.io)).exit;
809
810
  }
811
+ function checkArguments(tool, domain, forceRefresh) {
812
+ const declared = tool.inputSchema?.properties;
813
+ const wanted = { domain, format: "full", ...forceRefresh ? { force_refresh: true } : {} };
814
+ if (!declared) return wanted;
815
+ return Object.fromEntries(Object.entries(wanted).filter(([key]) => key in declared));
816
+ }
810
817
  async function runCheck(parsed, deps) {
811
818
  assertOptions(parsed, ["--format", "--out"], ["--force-refresh", "--force"]);
812
819
  if (parsed.positionals.length !== 2) throw new CliUsageError("Expected a check name and one domain");
@@ -821,15 +828,7 @@ async function runCheck(parsed, deps) {
821
828
  if (!tool || tool.annotations?.readOnlyHint !== true || tool.outputSchema === void 0) {
822
829
  throw new CliUsageError(`${toolName} is not a public read-only CheckResult tool`);
823
830
  }
824
- const result = await client.callTool(
825
- toolName,
826
- {
827
- domain,
828
- format: "full",
829
- ...parsed.flags.has("--force-refresh") ? { force_refresh: true } : {}
830
- },
831
- { readOnly: true }
832
- );
831
+ const result = await client.callTool(toolName, checkArguments(tool, domain, parsed.flags.has("--force-refresh")), { readOnly: true });
833
832
  const errorExit = classifyToolError(result, deps.io, toolName);
834
833
  if (errorExit !== void 0) return errorExit;
835
834
  const structured = CheckResultSchema.safeParse(requireStructured(result, toolName));