Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

MCP Gemini

PYPI · MCP-GEMINI-CRUNCHTOOLS · SCANNED SEP 20

MCP server for Google Gemini AI - text, image, video, research, and more

Available components

pypi · mcp-gemini-crunchtools 79
0 this week 79 Trust /100
Trust breakdown (7 categories)

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 native-build step with no shell scripting around it. View diagnostics → Pass
  • 3 of 38 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency35
  • 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: the license (GNU Affero General Public License v3 or later (AGPLv3+)) isn't a recognized OSI-approved license. See how to fix → Fail
  • Actively maintained (last published 201 days ago).Pass
  • Publishes a security disclosure policy (SECURITY.md).Pass
Schema Quality & AI Usability79
  • AI-judged instruction clarity (excellent).Pass
  • Tool/resource definitions use about 2600 tokens (~66/item across 39 items; 39 tools + 0 resources), lean.Pass
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management80
  • Stability observed for 24 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
  • Structured output schemas are declared (100% of tools); any adoption earns full credit.Pass
Tool Safety75
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • 0 of 2 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "gemini_delete_cache_tool" implies "delete" and declares no destructiveHint at all, which the MCP spec reads as destructive by default. See how to fix → Fail
  • An AI judge read all 40 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
Install

How do I install the MCP Gemini server?

MCP Gemini runs locally as a PyPI package, launched with uvx mcp-gemini-crunchtools. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

pypi · mcp-gemini-crunchtools

# add to Claude Code
claude mcp add crunchtools-gemini -- uvx mcp-gemini-crunchtools
// .cursor/mcp.json
{
  "mcpServers": {
    "crunchtools-gemini": {
      "command": "uvx",
      "args": [
        "mcp-gemini-crunchtools"
      ]
    }
  }
}
// .vscode/mcp.json
{
  "servers": {
    "crunchtools-gemini": {
      "command": "uvx",
      "args": [
        "mcp-gemini-crunchtools"
      ]
    }
  }
}
# add to Codex CLI
codex mcp add crunchtools-gemini -- uvx mcp-gemini-crunchtools
// opencode.json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "crunchtools-gemini": {
      "type": "local",
      "command": [
        "uvx",
        "mcp-gemini-crunchtools"
      ],
      "enabled": true
    }
  }
}
# add to OpenClaw
openclaw mcp add crunchtools-gemini --command uvx --arg mcp-gemini-crunchtools
# ~/.hermes/config.yaml
mcp_servers:
  crunchtools-gemini:
    command: "uvx"
    args: ["mcp-gemini-crunchtools"]
// ~/.netclaw/config/netclaw.json
{
  "McpServers": {
    "crunchtools-gemini": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "mcp-gemini-crunchtools"
      ]
    }
  }
}
# add to Vellum
assistant mcp add crunchtools-gemini -t stdio -c uvx -a mcp-gemini-crunchtools
// mcp.json
{
  "mcpServers": {
    "crunchtools-gemini": {
      "command": "uvx",
      "args": [
        "mcp-gemini-crunchtools"
      ]
    }
  }
}
Changelog

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 −3
    • Stability: pass → 0.80 functional
  • 19 Sept 26 +1
    • Stability: 0.97 → pass security
  • 17 Sept 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.

  • 15 Sept 26 +16
    • Malware scan: unverified → pass ▲ security
  • 14 Sept 26 −15
    • Malware scan: pass → unverified ▼ security
  • 12 Sept 26 −3
    • Stability: pass → 0.77 functional
  • 11 Sept 26 +1
    • Stability: 0.97 → pass security
  • 9 Sept 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 90 to 93. That category is still filling its 30-day observation window: 27 days of observed history at the previous scan, 28 at this one. The score rises as the window fills, whether or not the server changes.

Diagnostics

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 pypi/mcp-gemini-crunchtools@0.3.0

Provenance No attestation

The registry publishes no build provenance for this version, so there is nothing to verify.

Result No attestation
Ecosystem pypi

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 38 packages
Packages resolved 38
Stale 3
Tree resolution Complete

Background: SBOMs and build attestations, explained →

MCP tools · 39 exposed · ~2,456 tokens

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 →

Tool Tokens
gemini_analyze_code_tool ~69

Analyze code with Gemini.

NameTypeReqDescription
codestringyesThe code to analyze.
focusstring–Analysis focus (general, security, performance, bugs).
language––Programming language (auto-detected if not specified).
modelstring–Model to use.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_analyze_document_tool ~61

Analyze a document (PDF, DOCX, TXT, etc.).

NameTypeReqDescription
file_pathstringyesAbsolute path to the document.
modelstring–Model to use.
questionstring–Question or instruction about the document.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_analyze_image_tool ~63

Analyze and describe a local image file.

NameTypeReqDescription
image_pathstringyesAbsolute path to the image file.
modelstring–Model to use ('pro' or 'flash').
querystring–Question or instruction about the image.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_analyze_text_tool ~57

Analyze text with Gemini.

NameTypeReqDescription
analysis_typestring–Type of analysis (general, sentiment, tone, summary).
modelstring–Model to use.
textstringyesThe text to analyze.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_analyze_url_tool ~71

Analyze one or more URLs.

NameTypeReqDescription
modelstring–Model to use.
questionstring–Question or instruction about the URL content.
urlsarrayyesList of URLs to analyze (1-20).
use_google_searchboolean–Use Google Search grounding.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_brainstorm_tool ~66

Brainstorm ideas on a topic using Gemini.

NameTypeReqDescription
context––Additional context or constraints.
modelstring–Model to use.
num_ideasinteger–Number of ideas to generate.
topicstringyesThe topic to brainstorm about.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_check_research_tool ~39

Check the status of a deep research operation.

NameTypeReqDescription
research_idstringyesThe research ID from gemini_deep_research.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_check_video_tool ~36

Check the status of a video generation operation.

NameTypeReqDescription
operation_namestringyesThe operation name from gemini_generate_video.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_compare_urls_tool ~63

Compare two URLs.

NameTypeReqDescription
aspectstring–What to compare (content, design, sentiment, features).
modelstring–Model to use.
url1stringyesFirst URL.
url2stringyesSecond URL.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_continue_image_edit_tool ~56

Continue editing in an active image session.

NameTypeReqDescription
promptstringyesEditing instructions (e.g., "make the sky bluer").
session_idstringyesThe session ID from gemini_start_image_edit.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_count_tokens_tool ~45

Count the number of tokens in content.

NameTypeReqDescription
contentstringyesThe text content to count tokens for.
modelstring–Model to count tokens for.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_create_cache_tool ~112

Create a content cache for repeated queries. Upload content once and query it multiple times without re-sending.

NameTypeReqDescription
content––Text content to cache (alternative to file_path).
display_namestring–Human-readable name for the cache.
file_path––Path to a file to cache.
modelstring–Model to use.
system_instruction––System instruction for queries against this cache.
ttl_minutesinteger–Time-to-live in minutes (default 60).

Structured output declared, but exposes no named fields.

No examples provided.

gemini_deep_research_tool ~47

Start a deep research task using a specialized Gemini model. Performs multi-step web research to answer complex questions.

NameTypeReqDescription
querystringyesThe research question or topic.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_delete_cache_tool ~29

Delete a content cache.

NameTypeReqDescription
cache_namestringyesThe cache name to delete.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_dialogue_tool ~52

Generate multi-voice dialogue audio.

NameTypeReqDescription
textstringyesDialogue text with speaker labels.
voice1string–First voice name.
voice2string–Second voice name.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_end_image_edit_tool ~34

End an image editing session and free resources.

NameTypeReqDescription
session_idstringyesThe session ID to end.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_extract_from_url_tool ~75

Extract specific data from a URL.

NameTypeReqDescription
custom_fields––Comma-separated list of custom fields to extract.
data_typestring–Type of data (text, links, images, structured).
modelstring–Model to use.
urlstringyesURL to extract data from.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_extract_tables_tool ~56

Extract tables from a document.

NameTypeReqDescription
file_pathstringyesAbsolute path to the document.
modelstring–Model to use.
output_formatstring–Output format (markdown, csv, json).

Structured output declared, but exposes no named fields.

No examples provided.

gemini_extract_tool ~72

Extract structured data from text.

NameTypeReqDescription
custom_fields––Comma-separated list of custom fields to extract.
extract_typestring–Type (entities, dates, numbers, contacts, custom).
modelstring–Model to use.
textstringyesText to extract data from.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_generate_image_tool ~138

Generate an image from a text prompt using Gemini's native image model.

NameTypeReqDescription
aspect_ratiostring–Aspect ratio (1:1, 2:3, 3:2, 3:4, 4:3, 9:16, 16:9, etc).
image_sizestring–Output size (1K, 2K, 4K).
promptstringyesDescription of the image to generate.
style––Optional style modifier (photorealistic, cartoon, watercolor, etc).
use_google_searchboolean–Ground with Google Search results.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_generate_image_with_input_tool ~103

Generate/edit an image using a local image file as input. Upload a local image and give Gemini instructions for how to modify it. Use cases: add watermarks, change styles, composite images, etc.

NameTypeReqDescription
_aspect_ratiostring–Desired aspect ratio for the output (currently unused).
file_pathstringyesAbsolute path to a local image file.
promptstringyesInstructions for what to do with the input image.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_generate_video_tool ~65

Generate a video using Veo. Returns an operation ID to poll.

NameTypeReqDescription
_aspect_ratiostring–Video aspect ratio (currently unused).
negative_prompt––What to avoid in the video.
promptstringyesDescription of the video to generate.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_image_prompt_tool ~74

Help craft an effective image generation prompt.

NameTypeReqDescription
descriptionstringyesWhat you want the image to show.
modelstring–Model to use for prompt crafting.
mood––Desired mood or atmosphere.
style––Desired art style (photorealistic, cartoon, watercolor, etc).

Structured output declared, but exposes no named fields.

No examples provided.

gemini_imagen_generate_tool ~161

Generate images using Google Imagen 4 models. Models (sorted by price): - imagen-4.0-fast-generate-001 ($0.02/image, fastest) - imagen-4.0-generate-001 ($0.04/image, standard) - imagen-4.0-ultra-generate-001 ($0.06/image, highest quality)

NameTypeReqDescription
aspect_ratiostring–Aspect ratio (1:1, 3:4, 4:3, 9:16, 16:9).
modelstring–Imagen model to use.
number_of_imagesinteger–Number of images to generate (1-4).
promptstringyesDescription of the image to generate.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_list_caches_tool ~18

List all active content caches.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_list_image_sessions_tool ~19

List all active image editing sessions.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_list_voices_tool ~21

List available voices for text-to-speech.

Input schema present but exposes no named parameters.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_query_cache_tool ~47

Query content in a cache.

NameTypeReqDescription
cache_namestringyesThe cache name from gemini_create_cache.
questionstringyesQuestion to ask about the cached content.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_query_tool ~78

Query Gemini with a prompt. Supports Google Search grounding.

NameTypeReqDescription
modelstring–Model to use ('pro' or 'flash').
promptstringyesThe prompt to send to Gemini.
system_instruction––Optional system instruction to guide the response.
use_google_searchboolean–Ground response with Google Search results.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_research_followup_tool ~48

Ask a follow-up question about completed research.

NameTypeReqDescription
questionstringyesFollow-up question.
research_idstringyesThe research ID from a previous research operation.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_run_code_tool ~73

Execute code using Gemini's built-in code execution. Gemini writes and runs Python code to answer questions, perform calculations, or process data.

NameTypeReqDescription
data––Optional data to process.
modelstring–Model to use.
promptstringyesDescription of what code to write and run.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_search_tool ~40

Search the web using Gemini with Google Search grounding.

NameTypeReqDescription
modelstring–Model to use.
querystringyesThe search query.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_speak_tool ~60

Convert text to speech.

NameTypeReqDescription
textstringyesText to convert to speech.
voicestring–Voice name (Zephyr, Puck, Charon, Kore, Fenrir, Leda, Orus, Aoede).

Structured output declared, but exposes no named fields.

No examples provided.

gemini_start_image_edit_tool ~102

Start a multi-turn image editing session. Optionally provide an existing image file as the starting point. Without file_path, Gemini generates a new image from the prompt. Use gemini_continue_image_edit_tool to make further edits.

NameTypeReqDescription
file_path––Optional absolute path to a local image to start from.
promptstringyesInitial prompt (describe image or editing instructions).
use_google_searchboolean–Ground with Google Search results.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_structured_tool ~68

Get structured JSON output from Gemini.

NameTypeReqDescription
modelstring–Model to use.
promptstringyesThe prompt to generate structured output for.
schema––Optional JSON schema to constrain the output.
use_google_searchboolean–Ground response with Google Search.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_summarize_pdf_tool ~59

Summarize a PDF document.

NameTypeReqDescription
file_pathstringyesAbsolute path to the PDF.
modelstring–Model to use.
stylestring–Summary style (concise, detailed, executive).

Structured output declared, but exposes no named fields.

No examples provided.

gemini_summarize_tool ~70

Summarize content with Gemini.

NameTypeReqDescription
contentstringyesThe content to summarize.
formatstring–Output format (paragraph, bullets, outline).
lengthstring–Summary length (brief, moderate, detailed).
modelstring–Model to use.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_youtube_summary_tool ~58

Summarize a YouTube video.

NameTypeReqDescription
modelstring–Model to use.
stylestring–Summary style (concise, detailed, bullet_points).
urlstringyesYouTube video URL.

Structured output declared, but exposes no named fields.

No examples provided.

gemini_youtube_tool ~51

Analyze a YouTube video.

NameTypeReqDescription
modelstring–Model to use.
questionstring–Question or instruction about the video.
urlstringyesYouTube video URL.

Structured output declared, but exposes no named fields.

No examples provided.

Common questions

What is the MCP Gemini server?

MCP Gemini is listed in the public MCP registry as io.github.crunchtools/gemini. MCP server for Google Gemini AI - text, image, video, research, and more. This page covers its PyPI package (mcp-gemini-crunchtools).

Is the MCP Gemini server safe to use?

MCP Gemini scores 79 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 20 September 2026. 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 MCP Gemini server expose?

MCP Gemini exposes 39 tools: gemini_query_tool, gemini_brainstorm_tool, gemini_analyze_code_tool, gemini_analyze_text_tool, gemini_summarize_tool, and 34 more. Their descriptions and schemas cost roughly 2,456 tokens of context every time the server is loaded.

Is the MCP Gemini server still maintained?

MCP Gemini 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.