Before you start
- A Pro, Elite or Lifetime plan. Free accounts cannot create API keys (see pricing).
- An API key. Create one in Dashboard > API & MCP. It is shown once, so copy it straight into your client or a password manager.
- An MCP client, such as Claude Code, Claude Desktop, Cursor, VS Code, Windsurf, Codex or Gemini CLI.
Server address
https://miraresume.com/api/mcp| Transport | Streamable HTTP, stateless. Responses are JSON (no SSE stream; GET returns 405). |
|---|---|
| Server name | mira-resume |
| Protocol versions | 2025-11-25, 2025-06-18, 2025-03-26, 2024-11-05 |
| Capabilities | 11 tools and 2 prompts. Resources: none in v1 (resources/list returns an empty list). |
Authentication
Every request carries your API key. Keys start with mira_sk_ and act as you: anything the key does, it does on your account.
Authorization: Bearer mira_sk_...If your client cannot set an Authorization header, send x-api-key: mira_sk_... instead.
- A key is shown once when you create it. We store only a hash, so we cannot show it again; create a new one if you lose it.
- Revoke a key any time in Dashboard > API & MCP. It stops working on the next request.
- Pro accounts can hold 2 active keys, Elite 10. Use one key per client so you can revoke one without touching the others.
No OAuth sign-in yet
Mira Resume uses API keys only for now. Connectors that require an OAuth sign-in, such as ChatGPT connectors and claude.ai web connectors, cannot connect directly yet. ChatGPT works through a custom GPT with Actions on the REST API.Connect your client
Pick your client. Replace YOUR_API_KEY with your key, or follow the client's prompt where it asks for it.
- Create an API key in Mira Resume (Dashboard > API & MCP).
- Run the command below in your terminal. Add --scope user to use it in every project.
- Start Claude Code and type /mcp: "mira-resume" should show as connected.
- Ask: "List my resumes on Mira Resume."
claude mcp add --transport http mira-resume https://miraresume.com/api/mcp --header "Authorization: Bearer YOUR_API_KEY"- Install Node.js 18 or newer (Claude Desktop runs the small mcp-remote bridge with npx).
- In Claude Desktop open Settings > Developer > Edit Config.
- Paste the block below into claude_desktop_config.json (merge it if you already have mcpServers) and save.
- Quit and reopen Claude Desktop. The tools icon lists the Mira Resume tools.
{
"mcpServers": {
"mira-resume": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://miraresume.com/api/mcp",
"--header",
"Authorization:Bearer ${MIRA_API_KEY}"
],
"env": {
"MIRA_API_KEY": "YOUR_API_KEY"
}
}
}
}- The key sits in the env block, not in args, so it is not mangled by Windows argument parsing.
- macOS: ~/Library/Application Support/Claude/claude_desktop_config.json. Windows: %APPDATA%\Claude\claude_desktop_config.json.
- Open Cursor Settings > MCP > Add new global MCP server (this opens ~/.cursor/mcp.json).
- Paste the block below and save.
- Back in Settings > MCP, check that mira-resume shows a green dot and its tools.
- In the Agent chat ask: "Check my Data Analyst resume against this job: ..."
{
"mcpServers": {
"mira-resume": {
"url": "https://miraresume.com/api/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}- Create .vscode/mcp.json in your workspace (or run MCP: Open User Configuration for every workspace).
- Paste the block below and save. Click Start above the server name.
- VS Code asks for the key once and stores it securely.
- Open Copilot Chat in Agent mode and pick the Mira Resume tools from the tools menu.
{
"inputs": [
{
"type": "promptString",
"id": "mira-api-key",
"description": "Mira Resume API key",
"password": true
}
],
"servers": {
"mira-resume": {
"type": "http",
"url": "https://miraresume.com/api/mcp",
"headers": {
"Authorization": "Bearer ${input:mira-api-key}"
}
}
}
}- Open Windsurf Settings > Cascade > MCP Servers > View raw config.
- Paste the block below into mcp_config.json and save.
- Press Refresh in the MCP panel. The Mira Resume tools appear in Cascade.
{
"mcpServers": {
"mira-resume": {
"serverUrl": "https://miraresume.com/api/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}- Add the block below to ~/.codex/config.toml.
- Set the key in your shell profile: export MIRA_API_KEY=mira_sk_... (PowerShell: $env:MIRA_API_KEY = "mira_sk_...").
- Start codex and run /mcp to see the Mira Resume tools.
[mcp_servers.mira-resume]
url = "https://miraresume.com/api/mcp"
bearer_token_env_var = "MIRA_API_KEY"- Open ~/.gemini/settings.json (create it if needed).
- Paste the block below (merge with any existing mcpServers) and save.
- Start gemini and run /mcp to list the tools.
{
"mcpServers": {
"mira-resume": {
"httpUrl": "https://miraresume.com/api/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}- ChatGPT connects to remote MCP servers only with OAuth sign-in, which Mira Resume does not offer yet. Until then, use a custom GPT with Actions, which call the same tools through the REST API.
- In ChatGPT open Explore GPTs > Create > Configure > Create new action.
- Choose Import from URL and paste the OpenAPI address below.
- Under Authentication pick API Key, Auth Type Bearer, and paste your Mira Resume key.
- Save the GPT (keep it private: it acts as you) and ask it to list your resumes.
https://miraresume.com/api/v1/openapi.json- Native ChatGPT connector support (OAuth) is planned; this page will change when it ships.
Any MCP client that supports Streamable HTTP with custom headers can connect. Use the server URL and send the key in the Authorization header. The exact field names differ by client (for example url, serverUrl or httpUrl); the shape is:
{
"type": "http",
"url": "https://miraresume.com/api/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}Tools
All tools are available on Pro, Elite and Lifetime unless marked otherwise. Only one tool uses AI credits; everything else counts only toward your request limits.
| Tool | Group | Plan | Credits | Access |
|---|---|---|---|---|
get_account | Account | Pro and Elite | No credits | Read-only |
list_resumes | Resumes | Pro and Elite | No credits | Read-only |
get_resume | Resumes | Pro and Elite | No credits | Read-only |
create_resume | Resumes | Pro and Elite | No credits | Write |
update_resume | Resumes | Pro and Elite | No credits | Write (replaces data) |
check_resume_against_job | Resumes | Pro and Elite | No credits | Read-only |
list_applications | Job tracker | Pro and Elite | No credits | Read-only |
add_application | Job tracker | Pro and Elite | No credits | Write |
update_application | Job tracker | Pro and Elite | No credits | Write |
list_job_matches | Mira Scout | Pro and Elite | No credits | Read-only |
analyze_job_description | AI | Elite | 30 credits | Read-only |
get_accountGet account
Your plan, AI credits left this month, and today's API usage and limits.
Parameters
No parameters.
REST equivalent: GET /me
list_resumesList resumes
Lists your resumes with id, name, target role, last update and saved ATS score.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
limit | integer | Optional | Max resumes to return (default 50). |
REST equivalent: GET /resumes
get_resumeRead a resume
Returns one resume in full, as structured JSON or as Markdown.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
resume_id | string (uuid) | Required | The resume id (a UUID). |
format | "json" | "markdown" | Optional | "json" (default) returns structured sections; "markdown" returns readable text. |
REST equivalent: GET /resumes/{resume_id}
create_resumeCreate a resume
Creates a new resume, optionally with content. It appears in My Resumes and opens in the editor.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
name | string | Required | Name shown in My Resumes, e.g. "Data Analyst - Acme". |
target_role | string | Optional | The role this resume targets. |
resume | object | Optional | Initial content. Omit to start empty. |
REST equivalent: POST /resumes
update_resumeUpdate a resume
Replaces the sections you send (for example summary, skills or experience). Sections you leave out stay as they are.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
resume_id | string (uuid) | Required | The resume id (a UUID). |
name | string | Optional | - |
target_role | string | Optional | - |
resume | object | Required | Sections to replace. Sections you leave out are not changed. |
REST equivalent: PATCH /resumes/{resume_id}
check_resume_against_jobCheck resume against a job
Scores a resume against a job posting (0-100) with matched and missing skills. The same deterministic scorer as Mira Scout; no AI, no credits.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
resume_id | string (uuid) | Required | The resume id (a UUID). |
job_description | string | Required | The full job posting text. |
job_title | string | Optional | The job title, if it is not the first line of the posting. |
REST equivalent: POST /resumes/{resume_id}/match
list_applicationsList applications
Lists the applications in your Job Tracker, optionally by status.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
status | "wishlist" | "applied" | "interviewing" | "offered" | "rejected" | Optional | - |
limit | integer | Optional | Max rows (default 50). |
REST equivalent: GET /applications
add_applicationAdd an application
Adds a job to your Job Tracker (default status: wishlist).
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
company | string | Required | - |
role | string | Required | - |
job_url | string | Optional | - |
status | "wishlist" | "applied" | "interviewing" | "offered" | "rejected" | Optional | Default "wishlist". |
applied_date | string | Optional | YYYY-MM-DD. |
salary | string | Optional | - |
notes | string | Optional | - |
resume_id | string (uuid) | Optional | The resume version used for this application. |
REST equivalent: POST /applications
update_applicationUpdate an application
Changes an application's status, notes or next step.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
application_id | string (uuid) | Required | The application id (a UUID). |
status | "wishlist" | "applied" | "interviewing" | "offered" | "rejected" | Optional | - |
notes | string | Optional | - |
next_action | string | Optional | - |
next_action_date | string | Optional | YYYY-MM-DD. |
REST equivalent: PATCH /applications/{application_id}
list_job_matchesList Mira Scout matches
Jobs Mira Scout found for you, best first, with score, reasons and the link to apply.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
status | "new" | "saved" | "applied" | "dismissed" | Optional | Default "new". |
min_score | integer | Optional | - |
limit | integer | Optional | Max matches (default 25). |
REST equivalent: GET /scout/matches
analyze_job_descriptionAnalyze a job description (AI)
AI extraction of a posting's required and preferred skills, must-have keywords, seniority and responsibilities. Elite only; uses 30 credits, refunded if the AI fails.
Parameters
| Parameter | Type | Required | Description |
|---|---|---|---|
job_description | string | Required | The full job posting text. |
target_role | string | Optional | - |
REST equivalent: POST /ai/analyze-job
Prompts
Prompts are ready-made starting points. Clients that support MCP prompts show them in a menu or as slash commands.
tailor_resumeTailor a resume to a job
Reads your resume, checks it against a job posting and proposes edits before saving them.
| Argument | Required | Description |
|---|---|---|
resume_id | Yes | The resume to tailor (from list_resumes). |
job_description | Yes | The full job posting text. |
review_resumeReview a resume
A recruiter-style review of one resume: impact, keywords, length and ATS safety.
| Argument | Required | Description |
|---|---|---|
resume_id | Yes | The resume to review (from list_resumes). |
Limits
Limits are per account, across all your keys, and REST and MCP requests count together. The daily count resets at 00:00 UTC. If your plan changes to Free, your keys stop working until you upgrade again.
| Feature | Free | Pro | Elite |
|---|---|---|---|
| API and MCP access | No | Yes | Yes |
| Active API keys | - | 2 | 10 |
| Requests per day | - | 300 | 3,000 |
| Requests per minute | - | 30 | 90 |
| Resume, tracker and Mira Scout tools | - | Yes | Yes |
| Resume vs job match check | - | Free, unlimited within limits | Free, unlimited within limits |
| AI tools that use credits | - | No | Yes |
Errors
The MCP server and the REST API share these error codes. Over REST you get the HTTP status below with an error object; over MCP your assistant receives the same code and message.
| Code | Status | Meaning |
|---|---|---|
unauthorized | 401 | Missing, malformed, unknown or revoked API key. |
plan_required | 403 | Your plan does not include the API, or this tool needs Elite. |
scout_unavailable | 403 | Mira Scout is not available on your account yet. |
not_found | 404 | No resume or application with that id on your account. |
invalid_input | 400 | The body or a parameter failed validation; the message says which. |
rate_limited | 429 | Too many requests this minute. Retry after the Retry-After header. |
daily_limit | 429 | Today's request allowance is used up. Resets at 00:00 UTC. |
insufficient_credits | 402 | Not enough AI credits for this tool. |
ai_failed | 502 | The AI model did not return a usable answer. You were not charged. |
internal | 500 | Something went wrong on our side. |
Security
- Keys are stored as a SHA-256 hash with a short display prefix. They are never logged and never echoed back in errors.
- Every request is scoped to the account that owns the key. A key cannot read another user's data.
- The MCP server and the API use the key only: they do not read or set cookies.
- Inputs are validated against the same schemas shown on this page. Request bodies are capped at 1 MB and a resume at 200 KB of JSON.
- Writes go through the same checks as the resume editor, so a resume saved by your assistant opens normally in Mira Resume.
- Whatever your assistant reads (for example a resume) is processed by that assistant's provider under its own terms. Treat a key like a password and keep config files with keys out of shared repositories.
Troubleshooting
| Problem | Fix |
|---|---|
| 401 unauthorized | The key is missing, mistyped or revoked. Copy it again (it starts with mira_sk_), check there is exactly one space after "Bearer", or create a new key in Dashboard > API & MCP. |
| 403 plan_required | API access is part of Pro, Elite and Lifetime. If your plan lapsed to Free, the key stops working and starts again when you upgrade. The AI tool needs Elite. |
| 429 rate_limited or daily_limit | You hit the per-minute limit (wait for Retry-After seconds) or today's allowance (300 on Pro, 3,000 on Elite; resets at 00:00 UTC). Limits are shared across all your keys. |
| Claude Desktop shows the server as failed | Claude Desktop starts the mcp-remote bridge with npx, which needs Node.js 18 or newer. Install it, then quit and reopen Claude Desktop. |
| Windows: header with a space breaks | Some Windows launchers split arguments on spaces, so "Bearer <key>" arrives in two pieces. Use the env-variable form shown in the Claude Desktop config (Authorization:Bearer ${MIRA_API_KEY} with the key in env). |
| Tools do not show up | Save the config, then fully restart the client (quit, not just close the window). Check the JSON is valid: no trailing commas, and merge into an existing mcpServers block instead of adding a second one. |
| GET returns 405 | Expected. The server answers POST requests with JSON and does not open an SSE stream. Use a client that supports Streamable HTTP. |
| 402 insufficient_credits | The AI tool needs credits. Credits refill monthly; nothing else in the API uses them. |
Related guides
- Connect Mira Resume to Claude Code
- Connect Mira Resume to Claude Desktop
- Connect Mira Resume to Cursor
- Connect Mira Resume to VS Code
- Connect Mira Resume to Windsurf
- Connect Mira Resume to OpenAI Codex
- Connect Mira Resume to Gemini CLI
- Use Mira Resume in ChatGPT
- REST API reference and the OpenAPI spec
- Resume MCP server overview
- Mira Scout guide