Dark Search Protocol
NPM · DARK-SEARCH-MCP-SERVER · SCANNED SEP 20
161 dark data sources, 25+ domains, 6-step pipeline, Streamable HTTP + stdio
Available components
How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →
Supply Chain Security98
- No malware found by supply-chain analysis.Pass
- No known CVEs affecting this package version or its production dependencies.Pass
- No install/post-install scripts declared.Pass
- 31 of 95 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency19
- Repository check failed: the declared repository URL returned HTTP 404. See how to fix → View diagnostics → Fail
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- Clear OSI-approved license (MIT).Pass
- Actively maintained (last published 106 days ago).Pass
- Security-disclosure policy not yet verified: we couldn't inspect the source repository.Unverified
Schema Quality & AI Usability65
- AI-judged instruction clarity (excellent).Pass
- Context-footprint check failed: tool/resource definitions use about 1416 tokens (~283/item across 5 items; 5 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
- Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management90
- Stability observed for 27 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage100
- 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
- 100% of tool parameters carry a description.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 5 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 5 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
How do I install the Dark Search Protocol MCP server?
Dark Search Protocol runs locally as an npm package, launched with npx -y dark-search-mcp-server. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
npm · dark-search-mcp-server
claude mcp add parnasheps-eng-dark-search-protocol -- npx -y dark-search-mcp-server
{
"mcpServers": {
"parnasheps-eng-dark-search-protocol": {
"command": "npx",
"args": [
"-y",
"dark-search-mcp-server"
]
}
}
} {
"servers": {
"parnasheps-eng-dark-search-protocol": {
"command": "npx",
"args": [
"-y",
"dark-search-mcp-server"
]
}
}
} codex mcp add parnasheps-eng-dark-search-protocol -- npx -y dark-search-mcp-server
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"parnasheps-eng-dark-search-protocol": {
"type": "local",
"command": [
"npx",
"-y",
"dark-search-mcp-server"
],
"enabled": true
}
}
} openclaw mcp add parnasheps-eng-dark-search-protocol --command npx --arg -y --arg dark-search-mcp-server
mcp_servers:
parnasheps-eng-dark-search-protocol:
command: "npx"
args: ["-y", "dark-search-mcp-server"] {
"McpServers": {
"parnasheps-eng-dark-search-protocol": {
"Transport": "stdio",
"Command": "npx",
"Arguments": [
"-y",
"dark-search-mcp-server"
]
}
}
} assistant mcp add parnasheps-eng-dark-search-protocol -t stdio -c npx -a -y dark-search-mcp-server
{
"mcpServers": {
"parnasheps-eng-dark-search-protocol": {
"command": "npx",
"args": [
"-y",
"dark-search-mcp-server"
]
}
}
} Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.
- 20 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 87 to 90. That category is still filling its 30-day observation window: 26 days of observed history at the previous scan, 27 at this one. The score rises as the window fills, whether or not the server changes.
- 18 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 80 to 83. That category is still filling its 30-day observation window: 24 days of observed history at the previous scan, 25 at this one. The score rises as the window fills, whether or not the server changes.
- 17 Sept 26 −3
- Stability: pass → 0.80 functional
- 16 Sept 26 +1
- Stability: 0.97 → pass security
- 13 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 87 to 90. That category is still filling its 30-day observation window: 26 days of observed history at the previous scan, 27 at this one. The score rises as the window fills, whether or not the server changes.
- 11 Sept 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 80 to 83. That category is still filling its 30-day observation window: 24 days of observed history at the previous scan, 25 at this one. The score rises as the window fills, whether or not the server changes.
- 10 Sept 26 −3
- Stability: pass → 0.80 functional
- 9 Sept 26 +1
- Stability: 0.97 → pass security
Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.
Captured 20 Sept 2026 · Analysed npm/dark-search-mcp-server@0.8.0
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | npm |
Background: How many MCP packages publish verified provenance →
Dependencies 95 packages
| Packages resolved | 95 |
|---|---|
| Stale | 31 |
| Tree resolution | Complete |
Background: SBOMs and build attestations, explained →
The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →
add_source ~165
Submit a new dark data source to the Dark Search Protocol registry. USE THIS TOOL WHEN: - The user knows a specialized data source that should be included - Search results are missing a known source in the relevant domain - A user wants to contribute their proprietary/closed data to the network
| Name | Type | Req | Description |
|---|---|---|---|
| description | string | yes | What kind of data this source provides |
| layer | string | yes | Source layer: surface=publicly indexed, deep=not indexed but accessible, dark=proprietary/closed |
| name | string | yes | Human-readable name of the data source |
| tags | array | yes | Relevant tags in both English and Chinese, e.g., ['hvac', '暖通', 'repair'] |
| url | string | yes | URL of the data source |
No output schema declared.
No examples provided.
dark_query ~379
Query REAL structured data from specialized APIs that 99% of people don't know exist or how to use. Returns actual data, not just links. USE THIS TOOL WHEN: - You need real financial data (SEC filings, insider trades, company financials) - You need clinical trial data (drug trials, treatment outcomes, patient data) - You need academic paper metadata (citations, authors, open access links) - You need preprint research (latest ArXiv papers before peer review) - You need drug safety data (FDA adverse events, medical device problems) - You need global health statistics (WHO disease data, mortality rates) - You want structured JSON data, not web pages API SOURCES (all free, no authentication required): - SEC EDGAR: US public company financial filings (10-K, 10-Q, 8-K, insider trades) - ClinicalTrials.gov: Global clinical trial database (200,000+ studies) - OpenAlex: 250M+ academic papers with citations and open access links - ArXiv: Latest preprint papers in CS, physics, math, AI - FDA: Drug adverse events and medical device safety data - WHO GHO: Global health observatory data (disease, mortality, risk factors) Auto-detects which APIs to query based on your query content. Returns structured, parseable data.
| Name | Type | Req | Description |
|---|---|---|---|
| domain | string | – | Force a specific API domain: 'sec', 'clinicaltrials', 'openalex', 'arxiv', 'fda', 'who'. Leave empty for auto-detection. |
| max_results | number | – | Maximum number of results per API (default: 5) |
| query | string | yes | Your data query. Be specific: company names, drug names, research topics, disease names, etc. Supports English and Chinese. |
No output schema declared.
No examples provided.
dark_retrieve ~311
Retrieve ACTUAL content from internet archive sources — deleted pages, removed posts, and inaccessible content that mainstream search engines cannot find. USE THIS TOOL WHEN: - A web page returns 404, has been deleted, or is no longer accessible - You need to find what a page USED TO contain before it was changed or removed - You want to read deleted Stack Overflow answers or removed forum posts - You need archived historical content from Wayback Machine or Internet Archive - You want to verify what content existed at a URL at a specific point in time ARCHIVE ENGINES: - Wayback Machine: Retrieves archived snapshots of any URL (billions of pages archived since 1996) - StackPrinter: Retrieves deleted Stack Overflow/Stack Exchange content that is invisible to normal browsing - Internet Archive Search: Searches the full digital archive of books, papers, media, and datasets This tool returns REAL CONTENT, not just links. You can read the actual text of archived pages.
| Name | Type | Req | Description |
|---|---|---|---|
| max_results | number | – | Maximum number of results to return (default: 5) |
| query | string | yes | URL to retrieve from archives, or search query for archived content. Supports full URLs (https://example.com/page) or keyword queries. |
| type | string | – | Archive engine to use. 'auto' detects best engine based on query (default). 'wayback' for web archives. 'stackprinter' for Stack Overflow/Exchange. 'archive_org' for Internet Archive digital content. |
No output schema declared.
No examples provided.
dark_search ~430
Search deep and dark data sources NOT indexed by mainstream search engines (Google, Bing, etc.). USE THIS TOOL WHEN: - Web search returns insufficient, generic, or incomplete results - The query involves specialized, technical, niche, or industry-specific topics - You need data from forums, OEM manuals, proprietary databases, field reports, or other non-indexed sources - The user is asking about something obscure that mainstream search likely misses - You need real-world data from closed communities, professional forums, or invitation-only databases COVERED DOMAINS (122 built-in sources across 25 domains): Finance, Health/Medical, Food/Cooking, Shopping, Education, Law, Travel, Fashion, Copywriting, Psychology, Academic, HVAC, Cybersecurity, Embedded Systems, Industrial Control, Automotive, Real Estate, Career/Salary, AI/Programming, Gaming, Insurance, Pets, Parenting, Tax/Finance, Agriculture, Immigration and more. FEATURES (v0.6.0): - Semantic routing: Maps natural language queries to specialized dark data sources - Content fetching: Optionally crawls and returns actual content from matched sources - Cross-verification: Results validated across multiple sources - Confidence scoring: Each result rated HIGH/MID/LOW DO NOT USE FOR: - General knowledge questions (use web search instead) - Queries about popular/public topics with abundant mainstream coverage
| Name | Type | Req | Description |
|---|---|---|---|
| fetch_content | boolean | – | If true, attempt to fetch actual content from top result URLs (default: false). Use when you need real data, not just source suggestions. |
| layers | array | – | Which source layers to search. Default: all layers |
| max_results | number | – | Maximum number of results to return (default: 10) |
| min_confidence | number | – | Minimum confidence threshold. 0.5=high quality, 0.3=moderate, 0.15=exploratory (default: 0.15) |
| query | string | yes | The search query. Be specific and include technical terms for best results. Supports English and Chinese. |
No output schema declared.
No examples provided.
list_sources ~131
List all available dark data sources in the Dark Search Protocol registry. USE THIS TOOL WHEN: - You want to see what specialized data sources are available before searching - The user asks what kind of data DSP can access - You need to verify if a specific domain/industry has dark data coverage - You want to explore the 122 built-in sources across 25 domains
| Name | Type | Req | Description |
|---|---|---|---|
| domain | string | – | Filter sources by domain (e.g., 'finance', 'health', 'food', 'law', 'academic', 'hvac', 'tax') |
| layer | string | – | Filter by source layer |
No output schema declared.
No examples provided.
What is the Dark Search Protocol MCP server?
Dark Search Protocol is an MCP server listed in the public MCP registry as io.github.parnasheps-eng/dark-search-protocol. 161 dark data sources, 25+ domains, 6-step pipeline, Streamable HTTP + stdio. This page covers its npm package (dark-search-mcp-server).
Is the Dark Search Protocol MCP server safe to use?
Dark Search Protocol scores 76 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 20 September 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.
What tools does the Dark Search Protocol MCP server expose?
Dark Search Protocol exposes 5 tools: dark_search, dark_retrieve, dark_query, list_sources, add_source. Their descriptions and schemas cost roughly 1,416 tokens of context every time the server is loaded.
Is the Dark Search Protocol MCP server still maintained?
Dark Search Protocol is still listed as active in the MCP registry. We last reached this channel on 20 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.
What licence is the Dark Search Protocol MCP server under?
Dark Search Protocol declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.