mcpsurfr
OCI · GHCR.IO/NKAPILA6/MCPSURFR · 2 COMPONENTS
Local web search with multi-engine deep research and on-device re-ranking. No API keys.
Available components
Last scanned (8 hours ago)
Scores higher than 38% of OCI servers
Category Search and Retrieval
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 Security0
- No malware scan is available for this kind of package: the supply-chain vendors we use do not cover it. This is a permanent gap in our coverage, not a finding about the package.Unverified
- Known CVEs could not be checked: this artifact ships no SBOM or dependency manifest, so there is no dependency list to read.Unverified
- Install-script risk not yet assessed.Unverified
- Dependency health could not be checked: this artifact ships no SBOM or dependency manifest, so there is no dependency list to read.Unverified
Provenance & Transparency32
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
- License check failed: no license is declared. See how to fix → Fail
- Actively maintained (last published 6 days ago).Pass
- Disclosure check failed: no security disclosure policy was found in the source repository. See how to fix → Fail
Schema Quality & AI Usability74
- 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).Pass
- AI-judged instruction clarity (good).Pass
- Context-footprint check failed: tool/resource definitions use about 1715 tokens (~285/item across 6 items; 4 tools + 2 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 Management0
- Stability check failed: the tool surface changed between v1.1.1 and v1.2.3: 3 tool removals, 0 breaking changes, 1 additions. See how to fix → Fail
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
- Structured output schemas are declared (100% of tools); any adoption earns full credit.Pass
Tool Safety100
- No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
- We read all 4 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
- An AI judge read all 6 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
- Implements a current MCP spec version (2026-07-28).Pass
Unverified: 1 category
A category scored 0 because we could not verify it: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.
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.
- 29 Oct 26 −3
- Stability: 0.17 → unverified ▼ security
- Tool safety: pass → unverified ▼ security
- Stability: 0.17 → fail ▼ security
- Schema quality: 174 → 285 ▼ functional
- Tool coverage: 100 → unverified ▼ functional
- Capabilities: pass → unverified ▼ functional
- Schema quality: 100 → unverified ▼ functional
- Package version: v1.1.1 → v1.2.3 functional
- 28 Oct 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes.
- 26 Oct 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes.
- 24 Oct 26 +1
No change was recorded against any check on this day. Stability & Change Management went from 0 to 3. That category is still filling its 30-day observation window: 0 days of observed history at the previous scan, 1 at this one. The score rises as the window fills, whether or not the server changes.
- 23 Oct 26 38
First indexed and scored.
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 5 Nov 2026 · Analysed oci/ghcr.io/nkapila6/mcpsurfr@v1.2.3
Provenance No attestation
The registry publishes no build provenance for this version, so there is nothing to verify.
| Result | No attestation |
|---|---|
| Ecosystem | oci |
| Reason | No attestation published |
Background: How many MCP packages publish verified provenance →
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 →
deep_research Deep research ~440
Research a topic across several search terms and engines. Each term is searched on each backend, results are re-ranked locally by relevance, duplicates are removed, and the best pages are fetched as text. Backends: auto, brave, duckduckgo, google, grokipedia, mojeek, startpage, wikipedia, yahoo. - "auto" (default): ddgs metasearch over several engines, skipping blocked ones. Most reliable. - "brave": fast and usually available. - "wikipedia": exact article lookups, best with title-like terms. - "google", "startpage", "yahoo", "mojeek": often blocked. Empty engines are reported as warnings, and if every chosen engine is empty the terms are retried with "auto".
| Name | Type | Req | Description |
|---|---|---|---|
| backends | – | – | Backends to use. Defaults to ["auto"]. Unknown names are skipped with a warning. |
| exclude_domains | – | – | Never results from these sites. Example: ["pinterest.com"] |
| include_domains | – | – | Only results from these sites, subdomains included. Example: ["python.org", "stackoverflow.com"] |
| include_urls | boolean | – | Whether to include each page's source URL. |
| news | boolean | – | Search news articles instead of the web. Results then carry their published "date". News backends: auto, bing, duckduckgo, yahoo. Pair with recency for current events. |
| num_results_per_term | integer | – | Number of results to rank per search term per backend. |
| recency | – | – | Applied to every term. Only results from the past "day", "week", "month" or "year". |
| region | – | – | Region-language code like "us-en", "uk-en" or "de-de". Default: ddgs's own (us-en). |
| search_terms | array | yes | 2-5 specific, related search queries covering different angles. Example: ["postgres query planner", "postgres index types", "EXPLAIN ANALYZE tutorial"] |
| top_k_per_term | integer | – | Number of top-ranked results to keep per search term per backend. |
Structured output declared, but exposes no named fields.
No examples provided.
rag_search_ddgs Web search ~355
Search the web and return the most relevant pages as text, for when whole pages are wanted (search returns only the relevant passages). Results are re-ranked locally by semantic similarity to the query before the top pages are fetched.
| Name | Type | Req | Description |
|---|---|---|---|
| backend | string | – | Search backend, same names as deep_research. Default "auto" (ddgs metasearch over several engines, skipping blocked ones). Non-auto backends that return nothing are retried with… |
| exclude_domains | – | – | Never results from these sites. Example: ["pinterest.com"] |
| include_domains | – | – | Only results from these sites, subdomains included. Example: ["python.org", "stackoverflow.com"] |
| include_urls | boolean | – | Whether to include each page's source URL for citation. Default: True |
| news | boolean | – | Search news articles instead of the web. Results then carry their published "date". News backends: auto, bing, duckduckgo, yahoo. Pair with recency for current events. |
| num_results | integer | – | Number of initial search results to rank. More results provide better coverage but increase processing time. Default: 10 |
| query | string | yes | The search query. Use natural language questions or keywords. Example: "latest developments in quantum computing" |
| recency | – | – | Only results from the past "day", "week", "month" or "year". |
| region | – | – | Region-language code like "us-en", "uk-en" or "de-de". Default: ddgs's own (us-en). |
| top_k | integer | – | Number of top-ranked pages to fetch and return. Default: 5 |
Structured output declared, but exposes no named fields.
No examples provided.
read_url Read a page ~145
Read one page whose URL you already have: a link the user pasted, or one from earlier results. HTML or PDF. With a query, returns only the passages that answer it, the same way search does; without one, the page's main text (cut to the shared budget).
| Name | Type | Req | Description |
|---|---|---|---|
| keywords | – | – | Terms a good answer would contain, as for search. |
| max_passages | integer | – | Passages to return when there's a query. Default 8. |
| query | – | – | What you want from the page. Leave out to get the whole text. |
| url | string | yes | A public http(s) URL. Localhost and private network addresses are refused. |
Structured output declared, but exposes no named fields.
No examples provided.
search Search (passages) ~427
Search the web and return only the passages that answer the query, not whole pages: usually about a tenth of the text of rag_search_ddgs. Use this first. Tiers, cheapest first: 1. Search, then keep the top_pages results whose title/snippet match best. 2. Fetch those pages, split them into ~600-char passages, keyword-match them (BM25) against the query plus `keywords`. 3. Only if that leaves too many candidates, or matches nothing (a paraphrase), the local embedder ranks them by meaning. If nothing matched the keywords and the client can show questions, the user is asked whether they want the full pages instead.
| Name | Type | Req | Description |
|---|---|---|---|
| backend | string | – | Search backend, same names as deep_research. Default "auto". |
| exclude_domains | – | – | Never results from these sites. Example: ["pinterest.com"] |
| include_domains | – | – | Only results from these sites, subdomains included. Example: ["python.org", "stackoverflow.com"] |
| keywords | – | – | Terms a good answer would contain, including synonyms and exact phrases. Example for "how do I make postgres faster": ["index", "EXPLAIN ANALYZE", "query planner", "vacuum"]. Strongly improves result… |
| max_passages | integer | – | Passages to return. Default 8. |
| news | boolean | – | Search news articles instead of the web. Results then carry their published "date". News backends: auto, bing, duckduckgo, yahoo. Pair with recency for current events. |
| num_results | integer | – | Search results to consider. Default 10. |
| query | string | yes | The question or topic, in natural language. |
| recency | – | – | Only results from the past "day", "week", "month" or "year". |
| region | – | – | Region-language code like "us-en", "uk-en" or "de-de". Default: ddgs's own (us-en). |
| top_pages | integer | – | Pages to fetch and split. Default 5. |
Structured output declared, but exposes no named fields.
No examples provided.
How do I install the mcpsurfr MCP server?
mcpsurfr runs locally as a container image, launched with docker run --rm -i -e MCPSURFR_MAX_CHARS ghcr.io/nkapila6/mcpsurfr:v1.2.3. Ready-made configuration for Claude, Cursor, VS Code, Codex and 3 more is on this page, copied from each client's own documentation.
oci · ghcr.io/nkapila6/mcpsurfr
claude mcp add nkapila6-mcpsurfr -- docker run --rm -i -e MCPSURFR_MAX_CHARS ghcr.io/nkapila6/mcpsurfr:v1.2.3
This image reads MCPSURFR_MAX_CHARS. Set it in your client's env block for this server; docker run -e passes it through to the container.
What is the mcpsurfr MCP server?
mcpsurfr is an MCP server listed in the public MCP registry as io.github.nkapila6/mcpsurfr. Local web search with multi-engine deep research and on-device re-ranking. No API keys. This page covers its container image (ghcr.io/nkapila6/mcpsurfr).
Is the mcpsurfr MCP server safe to use?
mcpsurfr scores 38 out of 100 on VerifyMCP. 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 mcpsurfr MCP server expose?
mcpsurfr exposes 4 tools: rag_search_ddgs, search, read_url, deep_research. Their descriptions and schemas cost roughly 1,367 tokens of context every time the server is loaded.
Is the mcpsurfr MCP server still maintained?
mcpsurfr is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.