Command Palette

Search for a command to run...

UnylyUnyly
Browse all

Calibrd Agent

FreeNot checked

Would you make the shortlist? Score a job against your CV: fit, gaps, ATS score, likely questions.

GitHubEmbed

About

Would you make the shortlist? Score a job against your CV: fit, gaps, ATS score, likely questions.

README

Will a hiring manager shortlist you for this role? Calibrd scores one job posting against your CV and answers that before you spend an evening on the application. It shows where you fall short at your level, the questions each round will ask, how your CV lands with a recruiter and with an applicant tracking system, and it drafts the cover letter when you decide the job is worth it.

This repository is the plugin manifest only. The server runs at https://www.calibrd.com/mcp.

Listed on Smithery and Glama, and in the official MCP registry as com.calibrd/agent.

Install

Claude (claude.ai): Settings → Connectors → Add custom connector → URL https://www.calibrd.com/mcp.

Claude Code:

claude mcp add --transport http calibrd https://www.calibrd.com/mcp

Grok (grok.com / Grok Bot): grok.com/connectors → Custom → paste the URL.

Grok Build, Cursor, other MCP clients: copy .mcp.json into your project, or install this plugin from the xAI marketplace.

Give the tools a posting as text or as a link (public LinkedIn job pages, Greenhouse, Lever, Levels.fyi, careers pages). Connecting opens a Calibrd sign-in once. Reports are free, three a day. A pass removes the cap and unlocks every tool; payment is on calibrd.com, never in the chat.

Tools

Tool What it does Gate
calibrd_score_job Match score 0–100 and the gaps at your level Free
calibrd_report The full Calibrd report on one job Free, 3 a day
calibrd_review_cv Recruiter and ATS scores, what to fix Free
calibrd_cover_letter A draft from your CV for one job One free, then a pass
calibrd_status What you have on Calibrd Free
calibrd_get_pass A checkout link for a pass Free to ask

Running it over stdio

Most clients connect straight to https://www.calibrd.com/mcp and none of this is needed. bridge.mjs is here for the two cases that cannot: clients that speak only stdio, and directory scanners that need a process to start and introspect.

node ./bridge.mjs

No dependencies and no build step — it is one file on the Node standard library. It answers initialize and tools/list locally, so listing the tools needs no account and no network. Running one does: tools/call forwards to the hosted server with the bearer token in CALIBRD_ACCESS_TOKEN, and says so plainly when there isn't one. CALIBRD_MCP_URL overrides the endpoint.

The tool catalogue in bridge.mjs mirrors the hosted server. If the server gains or changes a tool, update the file — nothing enforces that automatically.

Your data

Calibrd Agent runs inside your assistant. Your CV, the job description and the report pass through your assistant's provider (Anthropic, OpenAI or xAI) under their terms. Calibrd itself keeps none of it: reports and CVs are generated and returned, not stored.

Full details: https://www.calibrd.com/agent and https://www.calibrd.com/privacy

License

MIT, for this manifest. Calibrd itself is a hosted service under its own terms.

from github.com/Calibrd/calibrd-agent

Installing Calibrd Agent

This server has no published package — it is built from source. Open the repository and follow its README.

▸ github.com/Calibrd/calibrd-agent

FAQ

Is Calibrd Agent MCP free?

Yes, Calibrd Agent MCP is free — one-click install via Unyly at no cost.

Does Calibrd Agent need an API key?

No, Calibrd Agent runs without API keys or environment variables.

Is Calibrd Agent hosted or self-hosted?

A hosted option is available: Unyly runs the server in the cloud, no local setup required.

How do I install Calibrd Agent in Claude Desktop, Claude Code or Cursor?

Open Calibrd Agent on unyly.org, pick your client tab (Claude Desktop, Claude Code, Cursor) and press Install — the config is generated automatically, no JSON editing.

Related MCPs

Compare Calibrd Agent with

Not sure what to pick?

Find your stack in 60 seconds

Author?

Embed badge for your README

Browse similar

All ai MCPs