mcpField
P
Ponisha Freelance

Ponisha Freelance

Local
github.comBusiness & Marketing0Updated 3d ago
RepositoryVisit website

About Ponisha Freelance

Ponisha Freelance is a read-only MCP server for exploring public data from Iran’s Ponisha freelance marketplace. It helps clients and agents assess projects, freelancers, bids, and skill-market trends without a Ponisha account.

Key features

  1. 01Search projects by skill, category, and budget
  2. 02Read project details and employer history
  3. 03Review bids on projects
  4. 04Search freelancer profiles and portfolios
  5. 05Summarize demand and budgets by skill

Use cases

  1. 01Find projects matching a skill and budget
  2. 02Assess competition on a project
  3. 03Review a freelancer’s work and feedback
  4. 04Check demand and typical budgets for a skill

MCP support

Exposes 14 read-only tools to MCP clients over stdio for public Ponisha data. It cannot submit bids, publish projects, or message users.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "sepehr071-ponisha": {
      "command": "uvx",
      "args": [
        "ponisha-mcp"
      ],
      "env": {
        "PONISHA_MCP_PROXY": "<PONISHA_MCP_PROXY>"
      }
    }
  }
}
Run commanduvx ponisha-mcp
Environment variables to set
PONISHA_MCP_PROXYOptional HTTP proxy URL for the Ponisha requests; system proxy settings are ignored. Also accepts a comma-separated list: one is used until it gets blocked or fails to connect, then the next.

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
pypiponisha-mcp0.2.0

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

ponisha-mcp: Ponisha (پونیشا) freelance marketplace MCP server for AI agents

ponisha-mcp is an unofficial, read-only MCP server that gives AI agents the public data of Ponisha (ponisha.ir), Iran's freelance marketplace.
Its 14 tools search projects (text, skills, category, budget, sort), show a project with the employer's track record, list the offers made on it, search freelancers, read profiles, reviews and portfolios, browse fixed-price service packages, and summarize demand, budgets and trends for a skill.
No API key or Ponisha account is needed. It works in Claude, Cursor, Copilot and any other MCP client.

For AI agents · Quick start · Tools · FAQ · فارسی

For AI agents

  • What: unofficial MCP server (model context protocol) for ponisha.ir: project search, project details, bids, freelancer search, public profiles, skills and categories, skill market summary.
  • Install: uvx ponisha-mcp or claude mcp add ponisha -- uvx ponisha-mcp. Transport: stdio. Needs uv (Python 3.10+).
  • Auth: none. Optional PONISHA_MCP_PROXY (HTTP proxy URL, or a comma-separated list); system proxy settings are ignored.
  • Access: read-only. It cannot log in, bid, post projects or message anyone, so no confirmation is needed before calling any tool.
  • Tools: 14, all prefixed pn_. Typical sequences: pn_skills -> pn_search_projects -> pn_project -> pn_bids; pn_search_freelancers -> pn_profile; pn_skills -> pn_market / pn_trend; pn_profile -> pn_reviews + pn_portfolios; pn_services -> pn_service.
  • Use when: the user wants freelance projects to bid on in Iran, what a kind of work pays, how crowded a project is, whether an employer is reliable, or who to hire for a skill.
  • Do not use when: the user wants to apply, message or post (read-only), or needs private contact data.
  • Output: compact JSON. Money is Toman. Times are Tehran time. Text is Persian user content.

Quick start

claude mcp add ponisha -- uvx ponisha-mcp

Any MCP client: command uvx, args ["ponisha-mcp"].

Tools

ToolWhat it does
pn_search_projectsProjects by text, skills, category, open/all, budget range, sort.
pn_projectOne project: brief, budget, deadline, employer track record, optional bid price sample.
pn_bidsOffers on a project: price, days, pitch, bidder level and rating (paged).
pn_search_freelancersFreelancers by text, skills, level, portfolio, ranked by score.
pn_profilePublic profile of a freelancer or employer.
pn_skillsSkill name -> slug/id (about 700 skills).
pn_categoriesTop-level categories, or the sub-categories of one.
pn_marketOpen projects, budget quartiles, competition and related skills for one skill.
pn_trendPosting rate, median budget and bids for a skill sampled across its whole history.
pn_reviewsReviews of a user with score breakdown, or recent reviews for a skill.
pn_portfoliosSearch about 500k portfolio items, or list one user's.
pn_servicesFixed-price service packages by category, price, rating.
pn_serviceOne package: deliverables, requirements, paid add-ons.
pn_user_projectsAn employer's or freelancer's finished/open projects.

FAQ

Is it official? No. It reads the same public API the ponisha.ir web app uses for anonymous visitors.

Why are bids slow? Ponisha serves the bids list slowly (seconds, up to 30 s on projects with 100+ bids). Keep limit small.

Why does the budget filter only look at one page? The site has no budget filter. The tool scans up to 50 projects per page and keeps those whose range fits.

فارسی

ponisha-mcp یک سرور MCP غیررسمی و فقط‌خواندنی برای پونیشا (ponisha.ir) است. با آن می‌توانید پروژه‌ها را با کلمه، مهارت، دسته و بودجه جستجو کنید، پیشنهادهای ثبت‌شده و سابقه کارفرما را ببینید، فریلنسرها و پروفایل‌های عمومی را جستجو کنید و وضعیت بازار یک مهارت را بسنجید. بدون کلید API و حساب کاربری. نصب: uvx ponisha-mcp.

مبلغ‌ها به تومان و زمان‌ها به وقت تهران (میلادی) هستند. این ابزار ثبت پیشنهاد، پیام یا ثبت پروژه انجام نمی‌دهد.

Author: Sepehr Radmard (سپهر رادمرد) · github.com/sepehr071 · linkedin.com/in/sepehr-radmard

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.