Borrowing it
Nothing to install: this file belongs to A1-x-Tech/mcp-google-custom-search. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/A1-x-Tech/mcp-google-custom-search/main/CLAUDE.mdgit clone --depth 1 https://github.com/A1-x-Tech/mcp-google-custom-searchWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/a1-x-tech/mcp-google-custom-search/claude-md)<a href="https://agentmods.dev/instructions/a1-x-tech/mcp-google-custom-search/claude-md"><img src="https://agentmods.dev/badge/instructions/a1-x-tech/mcp-google-custom-search/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/a1-x-tech/mcp-google-custom-search/claude-md"><img src="https://agentmods.dev/badge/instructions/a1-x-tech/mcp-google-custom-search/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.02472 | $0.02472 |
| Opus 5 | $0.01236 | $0.01236 |
| Sonnet 5 | $0.00494 | $0.00494 |
| Haiku 4.5 | $0.00247 | $0.00247 |
Grade A, and why
mcp-google-custom-search CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — mcp-google-custom-search
MCP server for the Google Custom Search JSON API (TypeScript, stdio). Read-only
end to end: search (web), search_images and the raw_request escape hatch
all hit the API's single GET endpoint customsearch/v1 on
https://customsearch.googleapis.com. Auth is a static Google Cloud API key
sent in the X-Goog-Api-Key header (no OAuth, no scopes, no token refresh) plus
a Programmable Search Engine id (cx) — configured as the default engine or
passed per call as engine_id. The API has no write endpoint; nothing here can
modify data.
Commands
npm run dev # run from source (tsx watch)
npm test # unit tests + dist smoke, no network
npm run typecheck # types for src + tests
npm run build # emit dist/
npm run smoke # live READ-ONLY check: one 1-result search (argv/GOOGLE_CUSTOM_SEARCH_SMOKE_QUERY)
Architecture
src/config.ts— env → config. Credentials:GOOGLE_CUSTOM_SEARCH_API_KEY(the key — required for anything to work) +GOOGLE_CUSTOM_SEARCH_ENGINE_ID(the defaultcx; addressing, not a credential — tools can override it per call); optionalGOOGLE_CUSTOM_SEARCH_API_BASE,GOOGLE_CUSTOM_SEARCH_TIMEOUT_MS,GOOGLE_CUSTOM_SEARCH_MAX_RETRIES. No credentials at all is NOT an error: the fields stayundefinedand the server starts degraded. An engine id without a key throwsConfigErrorincomplete_config(the key cannot come per call, so that install can never work). Also home toCredentialsError/MISSING_CREDENTIALS_MESSAGE(names both variables and the restart),EngineIdError/MISSING_ENGINE_ID_MESSAGE(missingcx: fix per call viaengine_idor via the env var + restart) andhasCredentials()(= the key is present).src/client.ts— all HTTP and all wire mapping.apiKey()throwsCredentialsErrorbefore any fetch;resolveEngineId()prefers the per-call override and throws a typedEngineIdError(naming both the env var and theengine_idparameter) when neither exists;request()resolves the path against the base and rejects foreign origins (SSRF guard), sends the key in theX-Goog-Api-Keyheader (never the URL), fills in the configuredcxwhen the path has none, enforces an AbortController timeout that also covers reading the body, retries 429 and 5xx/network errors with backoff honoringRetry-Afterin both RFC 9110 forms — delta-seconds and HTTP-date (every request is an idempotent GET — this API has no writes an ambiguous failure could duplicate) and throwsGoogleCustomSearchError(status, body).buildSearchQuery()/buildImageSearchQuery()map the normalized vocabulary to the wire parameters (lr=lang_<code>,cr=country<CC>,siteSearchFilteri/e,filter0/1, uppercaseimgSize,searchType=image);normalizeSearchResponse()reduces the envelope to query/totals/cursors/items with title-url-snippet-metadata (pagemap only on request).src/tools/search.ts—search, plus the sharedsearchInputFields()schema factory and thetoSearchParams()rename thatimages.tsreuses.src/tools/images.ts—search_images.src/tools/raw.ts—raw_request(GET only; the API has no other verbs).src/tools/util.ts—ok/fail, the four annotation presets (READ_ONLY/WRITE/UPDATE/DESTRUCTIVE; onlyREAD_ONLYis in use — see conventions) shared zod schema factories (querySchema,numSchema,startSchema,languageCodeSchema,countryCodeSchema,dateRestrictSchema,engineIdSchema) andcheckPageWindow(the cross-fieldstart + num - 1 <= 100guard the flat zod shapes cannot express — called by both search tools before the client, so an impossible window never burns a quota unit on a guaranteed 400).src/index.ts— wires everyregister*into the McpServer.loadConfigOrDegraded()catchesConfigError, pingsstartup_failed(fire-and-forget) and degrades the config to "no credentials"; an unconfigured start prependsUNCONFIGURED_PREFIX— plusConfiguration problem: <message>when a ConfigError was caught — to the initializeinstructions, andoninitializedsendsserver_startfor a configured install orunconfigured_start(with the reason) otherwise.src/telemetry.ts— anonymous usage pings (ids/names/versions only, never queries, results or arguments; fire-and-forget, must never block or throw; opt-outASKADS_TELEMETRY=0).server_startmeans "a usable install started";unconfigured_startis a degraded start andstartup_faileda malformed config caught at load — both carry areasonfrom a closed vocabulary (missing_credentials,incomplete_config) — never a variable's name or value.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 138 lines · 2,472 tokens per session scan A 21de1df4c239
mcp-google-custom-search CLAUDE.md is an instructions file published in the GitHub repository A1-x-Tech/mcp-google-custom-search (0 stars, last pushed 9d ago), licensed MIT. It adds 2,472 tokens to every session, about $0.0124 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.