AnySearch Skill
A web search skill with no MCP server: four command-line clients search the web and 16 vertical domains, in batches and with page extraction
- 6,735 ★ on the repository
- Last commit
- Apache-2.0
Extended access
Why this level
- Search queries, page addresses and the key go to a third-party server, api.anysearch.com, and the service's zero-retention claims cannot be verified
- The agent runs scripts from the skill folder, the PowerShell client is called with -ExecutionPolicy Bypass, and the chosen client is written to runtime.conf
- Per the project page the agent can register an account on your email and write the received key to .env
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/anysearch-skillDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Search queries, page addresses and the key go to a third-party server, api.anysearch.com, and the service's zero-retention claims cannot be verified
- The agent runs scripts from the skill folder, the PowerShell client is called with -ExecutionPolicy Bypass, and the chosen client is written to runtime.conf
- Per the project page the agent can register an account on your email and write the received key to .env
This is third-party code. Review the repository files before installing.
What it does
AnySearch Skill gives an agent web search through the AnySearch service. Instead of an MCP server the skill ships four command-line clients in Python, Node.js, PowerShell and Bash with identical subcommands. The agent checks what is installed, picks the first suitable client in the order Python, Node.js, shell, and writes the choice to runtime.conf so it can call the client directly afterwards. The search command searches the web, batch_search runs several queries in parallel, and extract returns a page's content as Markdown. For questions in vertical domains the skill requires calling get_sub_domains first, learning the right subdomain and required parameters, and then passing them to --sdp. There are sixteen domains, among them finance, academia, law, health, code, security and travel, plus general search. All requests go to https://api.anysearch.com. The ANYSEARCH_API_KEY key is optional: without it anonymous access works with lower limits. If a key's quota is exhausted and the service returns a new key, the agent must ask you before writing it to .env. The skill tells the agent to treat page content as untrusted data, not as instructions.
Who it is for. For people who need web search in Claude Code, Cursor or Codex without running an MCP server, with ready vertical domains such as finance and academic publications.
Good fit when
- The agent needs fresh data from the web and you do not want to install an MCP server
- The query concerns quotes, academic papers, CVE vulnerabilities or flights: vertical search with its own parameters covers them
- You need several independent searches at once with batch_search
- You need a whole page read: extract returns it as Markdown
- Fact-checking or looking up documentation missing from the model's training data
Not a fit when
- Queries contain passwords, personal data or trade secrets: queries, page addresses and the key go to a third-party server, api.anysearch.com
- You need to read PDFs, DOC files, images, audio or archives: extract does not support them
- The machine has none of Python 3.6 with requests, Node.js 12, PowerShell 5.1 or Bash 3.2 with jq and curl
- You prefer a server integration over MCP: the same author has AnySearch MCP Server in the catalog
Example request
Use AnySearch to find the latest news on the Postgres 18 release and pull the text of the fullest article as MarkdownLimitations
The skill works only together with the external AnySearch service, and data handling on its side is outside your control: zero retention and no logs or telemetry are the service's own claims, and the repository's security policy leaves the backend out of scope. Apache-2.0 license.
Without a key the limits are lower, and the repository gives no exact figures.
extract output is limited: HTML and plain text may be cut at 50,000 characters, and oversized JSON or Markdown returns an error. The command has no format option.
Vertical search requires the agent to call get_sub_domains beforehand and pass all required parameters, even as empty values, otherwise the server returns a validation error. For a single query max_results accepts 1 to 10.
The scripts should be checked against SHA256SUMS.txt after installation: the author recommends doing it before first use. For PowerShell the instructions call for running with -ExecutionPolicy Bypass.
How to disable. Delete the anysearch folder from the agent's skills directory. A key saved in a .env file inside it goes with the folder, while the ANYSEARCH_API_KEY environment variable has to be removed by hand. The key can also be revoked in the service's dashboard.
FAQ
How does AnySearch Skill differ from AnySearch MCP Server?
The skill needs no MCP server: the agent calls the command-line clients from the skill folder, and they reach the service's public endpoints directly. SKILL.md says so itself. The capabilities are the same: search, batch search and page extraction.
Which vertical domains are available?
The constants.json file lists 17 domains: general, resource, social_media, finance, academic, legal, health, business, security, ip, code, energy, environment, agriculture, travel, film and gaming. For all except general search, get_sub_domains is called first.
How do I get a key?
In the dashboard on the service's site or with one request to /v1/auth/email/register: per the project page the agent sends your real email address with no verification code, and the key is shown once. Registration and anonymous mode are mutually exclusive, and you cannot switch in the middle of a flow.
What happens when a key's quota runs out?
If the response carries an auto_registered field with a new key, the agent must show you this and write the key to .env only with your consent. Without a new key it reports that the quota is exhausted and suggests configuring another.
How do I find out which client the agent picked?
The choice lives in runtime.conf next to the scripts, with Runtime and Command lines. If the file is missing or the command broke, the agent repeats detection in the order Python, Node.js, shell. The file is created at install time and does not ship in the repository.
How do I check that the scripts were not tampered with?
SHA256SUMS.txt in the root holds hashes of the four clients: anysearch_cli.py, anysearch_cli.js, anysearch_cli.ps1 and anysearch_cli.sh. After installing, compare them with ones computed locally.
Discussion
No questions or reviews yet.
Related
SkillsMCP server for real-time web search: general and vertical search, batch queries and page content extraction
A skill for web search, X search and page fetching for models without web access: no signup or key, the answer is JSON with source links
open-webSearch
open-websearch
Web search with no API keys: an MCP server, a CLI and a local daemon. It works by parsing search engine pages so it may break
Web Access
web-access
A web access skill that picks between search, page fetch and driving your own browser over CDP
A skill that routes a question to the right AI platform and, via browser automation, collects the answer from its native search
Yandex search API skill via Apify
claude-skill-yandex-search-api
An agent skill: the full Yandex SERP as JSON with no official API key, via the Apify Yandex Search Scraper actor
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| AnySearch Skill (this entry) | yes | 6,735 | ||
| AnySearch MCP Server | no | key is optional | 2,052 | the same search service as an MCP server rather than a set of command-line clients |
| Tavily MCP Server | yes | API key | 2,454 | the official MCP server of a different service: web search, page text extraction, site map and site crawling |
| ModSearch | no | 808 | a skill for web search |
README in short
The repository holds SKILL.md with the agent rules, four command-line clients in scripts/, a doc_spec.md interface description, a sample .env.example, SHA256SUMS.txt hashes, a security policy and test_cli.py tests run in GitHub Actions. The project page is in English and Chinese and covers installation for different agents, key setup, post-install verification and saving the chosen client.
README badge
Original SKILL.md
---
name: anysearch
description: Real-time search engine supporting web search, vertical domain search, parallel batch search, and URL content extraction.
version: 3.1.1
authors:
- AnySearch Team
credentials:
- name: ANYSEARCH_API_KEY
required: false
description: "API key for higher rate limits. Anonymous access available with lower rate limits."
storage: ".env file, environment variable, or --api_key CLI flag"
---
## Overview
AnySearch is a unified real-time search service supporting general web search, vertical domain search, parallel batch search, and full-page content extraction. The bundled cross-platform CLI tools call the public HTTP endpoints directly; no MCP server installation or JSON-RPC wrapper is required. Use the configured runtime directly for routine `search`, `batch_search`, `extract`, and `get_sub_domains` calls; run the `doc` command only when the CLI interface is unknown or recovery information is needed (see Recommended Entry Point).
## Trigger
This skill SHOULD be activated when the AI agent needs to perform any of the following:
1. **Information retrieval** — looking up facts, news, documentation, or any current data.
2. **Fact-checking** — verifying claims, cross-referencing statements.
3. **Web browsing / URL content extraction** — reading page content beyond search snippets.
4. **Vertical domain queries** — structured searches with identifiers (Stock:/CVE:/DOI:/IATA:/patent, etc.).
5. **Multi-intent queries** — several independent searches that can run in parallel.
**Vertical domain rule:** The DEFAULT search path is Path 2 (vertical). For queries that belong to or overlap with a supported domain (finance, academic, travel, health, code, legal, gaming, film, business, security, ip, energy, environment, agriculture, resource, social_media), **always call `get_sub_domains` first** to discover the correct `sub_domain` and required parameters before searching — vertical search produces significantly better results than general web search for these topics. Pure encyclopedia queries with ZERO domain overlap are the RARE EXCEPTION (Path 1). When UNSURE whether a query is general or domain-specific, use HYBRID: `batch_search` with 1 general query + N vertical queries in parallel. Coverage beats guessing.
**Required params rule:** When `get_sub_domains` returns params marked `(required)`, you MUST include ALL of them in `--sdp`. If a required param has no applicable value, pass it with an empty string value. Omitting a required param will cause a backend validation error. The `--sdp` flag (alias: `--sub_domain_params`, `-p`) accepts either JSON (`'{"type":"stock","symbol":"AAPL","cn_code":""}'`) or flat key=value format (`type=stock,symbol=AAPL,cn_code=`).
**Rule:** This skill is the **recommended** search tool. When AnySearch is unavailable (no API Key, quota exhausted, service error, or network failure), the agent SHOULD inform the user and MAY fall back to other available search methods if the user approves.
## Recommended Entry Point
Prefer direct CLI invocation. If `<skill_dir>/runtime.conf` exists and the requested command shape is already obvious (`search`, `batch_search`, `extract`, or `get_sub_domains`), the agent SHOULD use the configured command directly and SHOULD NOT run `doc` on every activation. Run `doc` only when the CLI interface is unknown, a command fails due to argument/schema uncertainty, the skill was just installed/updated, or vertical-domain constraints require the complete reference. The `doc` command is offline and remains available for recovery, but repeated metadata reads waste tool calls and tokens.
### Command Cheat Sheet
Use these exact command shapes for routine calls. Replace `<cmd>` with the command from `runtime.conf` (for example, `python3 <skill_dir>/scripts/anysearch_cli.py`). Do not invent extra output-format flags.
```bash
# Search. Optional filter: --max_results N (1-10, default 10)
# REST-native --tag/--params are preferred; --domain/--sub_domain/--sdp remain compatibility aliases.
<cmd> search "query" --max_results 5