create_experience
Create a new work experience on LAPRAS(https://lapras.com). You can check the result at https://lapras.com/cv
LABSCO SUMMARY
The CV gets fixed at the moment you notice it is out of date.
Searching postings and editing the profile they match against are the same session, so a role that wants something you have done but never wrote down can be handled before you apply rather than after you are filtered out. Every write points back at the CV page on lapras.com, which is where you confirm what actually landed.
What it is
A LAPRAS client in two halves: a job search over postings, and write access to the profile those postings are matched against — work experiences, job summary, career aspirations and tech skills.
What you get
- Job search filtered by keyword, position, minimum annual salary, employment type, business type and work style
- The same search narrowed by programming language, framework, database and infrastructure, and by preferred conditions
- Full detail on a posting, including the URL the application goes through
- Work experiences listed, created, updated and deleted, each with organisation, position, a client-work flag and start and end month and year
- Your job summary read and rewritten
- Your stated career aspirations read and rewritten
- The tech skills list read and replaced
Requirements
A LAPRAS account with its API key in LAPRAS_API_KEY, and npx on your PATH.
Setup effort
One command — npx -y @lapras-inc/lapras-mcp-server
USE CASES
01
Finding roles that match the stack you actually work in
The search filters on language, framework, database and infrastructure as well as salary and work style, and the detail call carries the URL the application goes through.
02
Bringing a stale CV up to date before you apply
Read what is currently listed, add the work that never got written down, and correct the entries whose dates or scope have drifted — each write showing up on your LAPRAS CV page.
03
Rewriting how you describe yourself
The job summary and the statement of what you want next are separate fields, both readable and rewritable, so the framing can be adjusted for the roles the search is actually returning.
04
Making sure a skill you use every day shows up in matching
The skills list is what postings are matched against, so a technology you have used for two years but never listed can be added at the moment you notice the gap.
12 tools
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
delete_experience
Delete a work experience from LAPRAS(https://lapras.com). You can check the result at https://lapras.com/cv
get_experiences
Get work experiences on LAPRAS(https://lapras.com)
get_job_detail
Get detailed information about a specific job posting. You can apply for jobs through the URL provided in the response.
get_job_summary
Return the job summary on your LAPRAS profile.
get_tech_skill
Return the technologies, skills and qualifications listed on your LAPRAS profile.
get_want_to_do
Return the career aspirations recorded on your LAPRAS profile.
search_jobs
Search job by keyword, position, and minimum annual salary
update_experience
Update a work experience on LAPRAS(https://lapras.com). You can check the result at https://lapras.com/cv
update_job_summary
Replace the job summary on your LAPRAS profile. The result appears on your LAPRAS CV page.
update_tech_skill
Replace the list of technologies, skills and qualifications on your LAPRAS profile.
update_want_to_do
Replace the career aspirations on your LAPRAS profile. The result appears on your LAPRAS CV page.
BEFORE YOU INSTALL
What to know first
Tools you get
12 tools when we connected. The documentation calls the search tool search_job; the registered name is search_jobs. Only search_jobs and get_job_detail work without LAPRAS_API_KEY, and six tools, such as update_experience and delete_experience, change your live profile.
Requires
npx on PATH
Where it runs
All 4 documented clients run the server on your machine.
CONTEXT COST
Light Standing cost
Rated by the size of the 12 tool definitions that ride along on every turn, not by how many there are
INSTALLATION
Connect your client
Server setup
TERMINAL
npx -y @lapras-inc/lapras-mcp-serverMCP Clients
Claude Desktop, Cursor, Gemini CLI
JSON
{
"mcpServers": {
"lapras": {
"command": "npx",
"args": [
"-y",
"@lapras-inc/lapras-mcp-server"
],
"env": {
"LAPRAS_API_KEY": "<YOUR_LAPRAS_API_KEY>"
}
}
}
}Docker
DOCKER · JSON
{
"mcpServers": {
"lapras": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e",
"LAPRAS_API_KEY",
"laprascom/lapras-mcp-server:v0.4.0"
],
"env": {
"LAPRAS_API_KEY": "<YOUR_LAPRAS_API_KEY>"
}
}
}
}
