Groundhog
PYPI · GROUNDHOG-MCP
Search, read and research the web, with hidden text stripped by default and sources hashed
Available components
Last scanned (21 hours ago)
Scores higher than 98% of PyPI 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 Security100
- No malware found by supply-chain analysis.Pass
- No known CVEs affecting this package version or its production dependencies.Pass
- Runs hatchling.build at install time, a recognised build step with no custom scripting around it. View diagnostics → Pass
- 3 of 55 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency100
- Source repository is publicly reachable at the declared URL. View diagnostics → Pass
- Cryptographically verified build provenance (signed, bound to dmytrome/groundhog). View diagnostics → Pass
- Clear OSI-approved license (MIT).Pass
- Actively maintained (last published 8 days ago).Pass
- Publishes a security disclosure policy (SECURITY.md).Pass
Schema Quality & AI Usability76
- 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).Pass
- AI-judged instruction clarity (excellent).Pass
- Context-footprint check failed: tool/resource definitions use about 1354 tokens (~338/item across 4 items; 4 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 Management100
- No destabilizing schema changes in the last 30 days.Pass
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 5 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
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.
- 31 Oct 26 +1
- Security disclosure: unverified → pass ▲ functional
- 30 Oct 26 −1
- Security disclosure: pass → unverified ▼ functional
- 29 Oct 26 +15
- Malware scan: unverified → pass ▲ security
- 26 Oct 26 +1
- Stability: 0.97 → pass security
- Package version: 0.20.1 → 0.22.0 functional
- 25 Oct 26 −1
- Stability: pass → 0.97 functional
- Package version: 0.20.0 → 0.20.1 functional
- 24 Oct 26 0
- Schema quality: 284 → 329 ▼ functional
- Package version: 0.19.1 → 0.20.0 functional
- 23 Oct 26 0
- Package version: 0.18.2 → 0.19.1 functional
- Package version: 0.18.2 → 0.19.0 functional
- 22 Oct 26 −15
- Stability: pass → unverified ▼ security
- Tool safety: pass → unverified ▼ security
- Malware scan: pass → unverified ▼ security
- Capabilities: pass → unverified ▼ functional
- Tool coverage: 100 → unverified ▼ functional
- Schema quality: 100 → unverified ▼ functional
- Package version: 0.18.0 → 0.18.2 functional
- Package version: 0.18.0 → 0.18.1 functional
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 4 Nov 2026 · Analysed pypi/groundhog-mcp@0.22.0
Provenance Verified
A signed build attestation was found and verified, binding this exact artifact to the source repository it claims to come from.
| Result | Verified |
|---|---|
| Ecosystem | pypi |
| Reason | Verified |
| Discovered via | Registry attestation endpoint |
| Source repo | dmytrome/groundhog |
| Certificate issuer | https://token.actions.githubusercontent.com |
| Certificate SAN | https://github.com/dmytrome/groundhog/.github/workflows/release.yml@refs/tags/v0.22.0 |
| Rekor log index | 3482273215 |
| Predicate type | PyPI publish attestation https://docs.pypi.org/attestations/publish/v1 |
| Subject digest | sha256:34c9c67e1edcff2dd86d977a623de5e93c680db4fb6b0a26ddf6aa6462578662 |
Background: How many MCP packages publish verified provenance →
Install scripts 1 script
| Hook | Tier | Command |
|---|---|---|
| build_backend | allowlisted | hatchling.build |
Background: Why install scripts are a supply-chain risk →
Dependencies 55 packages
| Packages resolved | 55 |
|---|---|
| Stale | 2 |
| No linked repository | 1 |
| 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 →
read_url Read a URL ~360
Fetch one web page through the stealth browser and return clean, grounded content with provenance. Hidden text injected for models but invisible to humans is stripped by default and reported in `threats`. Use this to ground answers in live web content, including sites that block plain fetchers. Reads a URL you already have: use `search` to find URLs, or `research` to search and read in one call. Fetches are rate limited per domain (5s apart by default), so several pages from one site are not instant.
| Name | Type | Req | Description |
|---|---|---|---|
| format | string | – | 'markdown' extracts the article; 'text' returns the page's rendered text. |
| include_hidden | boolean | – | Keep text that is invisible to a human reader. It is reported in `threats` either way; this only controls whether it stays in the content. |
| max_tokens | – | – | Token budget for the content. Omit it, or pass a value below 1, to use the server's GROUNDHOG_MAX_TOKENS (20000 by default). |
| query | – | – | When set, the content is built from the passages most relevant to it, and `matches` lists each one's heading, offset and score for citation; it carries no text. If nothing matches, `matches` is empty… |
| url | string | yes | Absolute http(s) URL. Private and loopback addresses are refused. |
| Name | Type | Req | Description |
|---|---|---|---|
| fetched_at | string | yes | – |
| final_url | string | yes | – |
| http_status | – | yes | – |
| markdown | string | yes | – |
| matches | array | yes | – |
| provenance | – | yes | – |
| set_aside | integer | yes | – |
| status | string | yes | – |
| threats | array | yes | – |
| title | string | yes | – |
| truncated | boolean | yes | – |
| url | string | yes | – |
No examples provided.
research Research the web ~470
Search the web and return ranked passages drawn from several sources. One call does what `search` + repeated `read_url` would: finds pages, reads them through the stealth browser, and returns the passages most relevant to `query` — each attributed to its source, with that source's provenance receipt and any stripped injection payloads. A source that fails is reported in `sources` rather than failing the whole call. Prefer `read_url` when you already have the URL, and `search` when you only want links. This reads `max_sources` pages, rate limited per domain, so it is the slowest of the three and the one to avoid for a single known page.
| Name | Type | Req | Description |
|---|---|---|---|
| aim | – | – | What the evidence must show, in at most 200 characters: the authority, artifact and version sought. Echoed in the receipt; it does not change the search or the ranking. |
| exclude_domains | – | – | Never read pages on these domains (at most 50). |
| follow | – | – | Read up to 8 of the `candidates` an earlier call offered instead of searching: each entry is a candidate's `token` and a `why` (at most 160 characters) saying what it should settle. Pass `query` and… |
| include_domains | – | – | Read only pages on these domains or their subdomains (at most 8), e.g. a standards body or the vendor's docs. Several pages from one domain may be read. |
| max_sources | integer | – | How many pages to read. Values outside 1-10 are clamped rather than rejected. Each source is a full page fetch, so this is the main cost and latency control. |
| max_tokens | – | – | Token budget for the returned passages. Omit it, or pass a value below 1, to use the server's GROUNDHOG_MAX_TOKENS (20000 by default). |
| published_after | – | – | YYYY-MM-DD. Leave out pages dated before this day. A page with no date is kept and counted in the receipt's `unknown_dates`. |
| query | string | yes | The question to research. Passages are ranked against it. |
| Name | Type | Req | Description |
|---|---|---|---|
| backend | string | yes | – |
| candidates | array | yes | – |
| followed | array | yes | – |
| notes | array | yes | – |
| omitted_safety | – | yes | – |
| passages | array | yes | – |
| query | string | yes | – |
| receipt | – | yes | – |
| sources | array | yes | – |
| truncated | boolean | yes | – |
No examples provided.
search Web search ~174
Search the web and return ranked hits (title, url, snippet, engine). Not for reading: it returns links, never page content. Use `read_url` for one known URL, or `research` when you want the answer rather than the links. Use this to find pages, then pass the URLs you want to `read_url` for safe, grounded content. Results come from a self-hosted SearXNG instance when `SEARXNG_URL` is set, otherwise from a search page rendered through the stealth browser. Hits are links only — nothing is fetched until you ask.
| Name | Type | Req | Description |
|---|---|---|---|
| limit | integer | – | How many hits to return. Values outside 1-25 are clamped rather than rejected. |
| query | string | yes | What to search for. Must not be empty. |
| Name | Type | Req | Description |
|---|---|---|---|
| backend | string | yes | – |
| hits | array | yes | – |
| query | string | yes | – |
No examples provided.
status Browser status ~49
Check whether Groundhog can reach the stealth browser. Call this to diagnose setup before fetching: if `browser_reachable` is false, follow `hint` to start the browser, then retry.
Input schema present but exposes no named parameters.
| Name | Type | Req | Description |
|---|---|---|---|
| browser_reachable | boolean | yes | – |
| cdp_url | string | yes | – |
| hint | – | yes | – |
No examples provided.
How do I install the Groundhog MCP server?
Groundhog runs locally as a PyPI package, launched with uvx groundhog-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.
pypi · groundhog-mcp
claude mcp add dmytrome-groundhog-mcp -- uvx groundhog-mcp
What is the Groundhog MCP server?
Groundhog is an MCP server listed in the public MCP registry as io.github.dmytrome/groundhog-mcp. Search, read and research the web, with hidden text stripped by default and sources hashed. This page covers its PyPI package (groundhog-mcp).
Is the Groundhog MCP server safe to use?
Groundhog scores 96 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 4 November 2026. Its build provenance is signed and verified. 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 Groundhog MCP server expose?
Groundhog exposes 4 tools: read_url, research, search, status. Their descriptions and schemas cost roughly 1,053 tokens of context every time the server is loaded.
Is the Groundhog MCP server still maintained?
Groundhog is still listed as active in the MCP registry. We last reached this channel on 4 November 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 Groundhog MCP server under?
Groundhog declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.