mcpField
S
Solvecha

Solvecha

Remote
solvecha.comBrowser Automation0Updated today

About Solvecha

Solvecha provides captcha-solving API tools and agent skills for AI agents that need to handle image and text captchas, GeeTest sliders, or FunCAPTCHA. It offers a remote MCP endpoint alongside Python clients and framework tools.

Key features

  1. 01Solve reCAPTCHA image grids
  2. 02Solve text captchas
  3. 03Handle GeeTest sliders
  4. 04Handle FunCAPTCHA
  5. 05Python API client and framework tools
  6. 06Agent skills for multiple environments

Use cases

  1. 01Automating captcha handling in AI agent workflows
  2. 02Adding captcha solving to Python framework projects

MCP support

A remote MCP server at https://api.solvecha.com/mcp exposes Solvecha captcha-solving capabilities to MCP clients. An API key is required.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://api.solvecha.com/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

All endpoints
Streamable HTTPhttps://api.solvecha.com/mcpAuth required

Tools

Details

From the project README

Solvecha for AI agents

Skills and framework tools for Solvecha, a captcha solving API (reCAPTCHA image grids, text captchas, GeeTest sliders, FunCAPTCHA).

PathWhat
skills/solvecha/Agent skill (SKILL.md, scripts in Python and Node.js, references). Works in Claude Code, Claude Desktop, OpenClaw and Hermes Agent.
python/solvecha Python package: API client and tools for OpenAI, LangChain, CrewAI, LlamaIndex and Browser Use.

The fastest way for most agents is the remote MCP server: https://api.solvecha.com/mcp (header Authorization: Bearer <key>). See https://solvecha.com/docs/ai-agents.

You need an API key from https://solvecha.com/dashboard. The free tier has 100 successful solves a day. Never commit a key. The skill and the package read SOLVECHA_API_KEY.

Install the skill

# Claude Code (personal)
cp -r skills/solvecha ~/.claude/skills/solvecha

# OpenClaw
cp -r skills/solvecha ~/.openclaw/skills/solvecha
# or from Git: openclaw skills install git:<owner>/<repo>

# Hermes Agent
hermes skills tap add Solvecha/solvecha-skills
hermes skills install solvecha
# or by hand (the folder under skills/ is the install name): cp -r skills/solvecha ~/.hermes/skills/solvecha

One command for the agents on your machine (Claude Code, Cursor, Codex and others), through skills.sh:

npx skills add Solvecha/solvecha-skills

Claude Desktop and claude.ai: zip the solvecha folder (the folder itself at the top of the zip) and upload it under Customize, Skills.

Validate with the official checker: pip install "git+https://github.com/agentskills/agentskills#subdirectory=skills-ref" && skills-ref validate skills/solvecha.

Python package

pip install solvecha
pip install "solvecha[langchain]"   # or openai, crewai, llamaindex, browser-use

On PyPI: https://pypi.org/project/solvecha/. See python/README.md for the client and the framework tools.

Releases

A published GitHub release uploads the Python package to PyPI (.github/workflows/publish.yml, Trusted Publishing, no stored token). The skill is not on npm or ClawHub; it installs from this repository with npx skills add and the Hermes tap.

MIT licensed. Solvecha is not affiliated with CloakHQ, NopeCHA, 2Captcha or CapSolver.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.