@qmediat.io/gemini-code-context-mcp

MCPattested
v1.21.0Quantum Media TechnologiesMITUpdated 9d agonpmGitHub

MCP server that gives Claude Code and other MCP hosts long-context memory of a codebase through Gemini's 1M-token window: `ask`/`code` send the scanned workspace with implicit caching by default or an explicit Context Cache on request; `ask_agentic` reads

Works in
ClaudeCursorCopilotGemini

Inferred from the transports this listing declares (stdio). A client not listed here hasn’t been ruled out — it just isn’t something Forge can confirm.

Attested build
A verified provenance attestation binds this artifact to the listed repository. Nobody has claimed the listing yet — this proves where the code was built, not who stands behind it.
306Downloads/wk
1Forks
9d agoLast update
Reads these credentials
  • GEMINI_API_KEYAPI keyoptional

    Fallback Tier 3 auth. Your Gemini API key. Emits a warning at startup recommending you move it to the credentials profile via the init command.

Declared by the author in the official MCP registry. Forge does not store, broker, or ever see these values — the config below is scaffolded with placeholders you fill in locally.

Package
AuthorQuantum Media Technologies
LicenseMIT
Version1.21.0
Sourcenpm+mcp-registry
Trust Status
A
85/100Trusted
✓Listed in Forge index+10/10
✓Identity verified · attested build+20/20
—Ed25519 publish signature+0/5
→ Included automatically when the publisher runs `forge publish`
—Domain verification+0/5
→ Publisher: host /.well-known/forge.json on the package homepage with { "publisher": "<github-login>" }
✓npm Trusted Publishing (Sigstore)+5/5
—npm maintainer match+0/5
→ Publisher: add the verified GitHub login to the npm package's maintainers (npm owner add <login>)
✓CVE scan · clean+30/30
✓Static analysis · clean+20/20
Paste into Claude Code, Cursor, or any AI assistant to fix all gaps
StatusIdentity verified
PublisherUnverified
SignatureUnsigned
Domain—
Provenance✓ Sigstore-verified · fe3a14f
Dependencies✓ 60 resolved+ · none vulnerable
Tool surface10 tools · none privileged
Security scan✓ Cleanv1.21.0 · todayHow well does this scan work?
EvalsNone
IndexedOct 11, 2026

Verification confirms publisher identity (repo ownership), not code safety. The security scan covers known CVEs and suspicious install scripts.

Tools

10 tools · none privileged
Statically extracted from the published packagev1.21.0 · 23h ago

Read out of the source npm actually ships, at scan time. The package was never executed. Tools registered dynamically at runtime, or hidden inside bundled or minified code, can be missed — so this is a floor on the tool surface, not a complete census of it.

list_directoryList the immediate children (files + subdirectories) of a workspace-relative directory path. Does NOT recurse. Excluded directories (node_modules, .git, .next, etc.) and denylisted filenames (lockfiles, .env, etc.) are silently filtered out.

List the immediate children (files + subdirectories) of a workspace-relative directory path. Does NOT recurse. Excluded directories (node_modules, .git, .next, etc.) and denylisted filenames (lockfiles, .env, etc.) are silently filtered out.

No input schema was published for this tool.

find_filesRecursive glob search for files under the workspace root. Supports `*` (match any character except `/`) and `**` (match any character including `/`). Examples: `**/*.ts`, `src/**/index.*`, `README.md`. Respects the same default excludes as list_directory.

Recursive glob search for files under the workspace root. Supports `*` (match any character except `/`) and `**` (match any character including `/`). Examples: `**/*.ts`, `src/**/index.*`, `README.md`. Respects the same default excludes as list_directory.

No input schema was published for this tool.

read_fileRead a file from the workspace, optionally a line-range slice. Responses are hard-capped at 200 KB; if the file is larger you get a head-truncated slice with `truncated: true` and `totalLines` metadata. Call again with `startLine` / `endLine` to page through. Rejects binary / non-source extensions.

Read a file from the workspace, optionally a line-range slice. Responses are hard-capped at 200 KB; if the file is larger you get a head-truncated slice with `truncated: true` and `totalLines` metadata. Call again with `startLine` / `endLine` to page through. Rejects binary / non-source extensions.

No input schema was published for this tool.

grepRegular-expression search through source files in the workspace. Pattern is compiled as a JavaScript RegExp (standard metacharacters). Each match line is capped at 500 characters. At most 100 matches per call.

Regular-expression search through source files in the workspace. Pattern is compiled as a JavaScript RegExp (standard metacharacters). Each match line is capped at 500 characters. At most 100 matches per call.

No input schema was published for this tool.

ask_agenticNo description published

This tool published no description. Forge does not invent one.

askNo description published

This tool published no description. Forge does not invent one.

clearDelete the cached context and manifest entries for a workspace. The Gemini-side cache is also released. Useful before switching projects or when you want to reset state deliberately.

Delete the cached context and manifest entries for a workspace. The Gemini-side cache is also released. Useful before switching projects or when you want to reset state deliberately.

No input schema was published for this tool.

codeDelegate a coding task to Gemini using its native thinking budget and (optional) code execution. Returns structured edits in OLD/NEW format that Claude Code can apply via its Edit tool, plus a brief rationale.

Delegate a coding task to Gemini using its native thinking budget and (optional) code execution. Returns structured edits in OLD/NEW format that Claude Code can apply via its Edit tool, plus a brief rationale.

No input schema was published for this tool.

reindexClear the current Gemini Context Cache for a workspace and rescan the file tree. The next ask/code call will upload changed files and build a new cache. Use after a large refactor when hash-based invalidation would be slower than a full rebuild.

Clear the current Gemini Context Cache for a workspace and rescan the file tree. The next ask/code call will upload changed files and build a new cache. Use after a large refactor when hash-based invalidation would be slower than a full rebuild.

No input schema was published for this tool.

statusInspect the cache state for a workspace: file count, cache ID, TTL remaining, cumulative cost, and savings from cache hits. Also lists models available to the current API key.

Inspect the cache state for a workspace: file count, cache ID, TTL remaining, cumulative cost, and savings from cache hits. Also lists models available to the current API key.

No input schema was published for this tool.

8 of 10 tools published a description.

Tool names and descriptions are written by the publisher and shown verbatim as inert text. They are the strings an MCP client passes to a model, so Forge scans them for prompt-injection patterns — any finding appears with the security scan above. “Privileged” is a keyword match on the tool name, not an audit of what the tool does: a benign-sounding name can still do anything.

About

MCP server that gives Claude Code and other MCP hosts long-context memory of a codebase through Gemini's 1M-token window: `ask`/`code` send the scanned workspace with implicit caching by default or an explicit Context Cache on request; `ask_agentic` reads

Keywords
mcpmodel-context-protocolgeminiclaude-codecontext-cachecode-reviewcodebase-qalong-context
Alternatives
Comparing tool surfaces…

Dependency tree

What one Forge scan resolved from npm metadata on 2026-10-15 — observed resolution, not a publisher declaration.

60 packages resolved · 5 direct · none carrying advisories Resolution stops at depth 4 and 60 packages.

The crawl stopped at the 60-package limit. The rest of the tree was never resolved.

36 more resolved packages are not drawn here (display cap: 24). Every dependency carrying an advisory is drawn regardless of the cap. Full inventory (CycloneDX SBOM)

Declared but not resolved

60 declared dependencies never landed in the tree. They are missing from Forge's resolution, not from the package.

+48 more not listed. The counts by reason above cover all of them.

Not followed: peerDependencies. This tree covers runtime dependencies only, so anything those pull in was never resolved.