Company Jobs Finder in Pi
Open jobs at named companies. The MCP server is https://jobs.openkrill.app/mcp. No key. The steps below are only for Pi. A call to find_company_jobs was checked against that server on 2026-10-01.
What this server answers
Ask about open jobs at the companies you care about and get the current list straight from their public careers boards.
Ask "what jobs is Stripe hiring for?", "is Notion hiring designers?", "remote backend jobs at Ramp, Linear and Supabase" or "which Spotify jobs list a salary?".
Company Jobs Finder reads the public job boards that companies publish on Greenhouse, Lever, Ashby, Workable and SmartRecruiters.
It can narrow the list by keyword, location, remote work, department, posting date and posted salary, and it links each job to its application page.
It looks up companies by name, up to five at a time. Companies whose careers pages run on other systems are reported as not found.
It can also help you pick and pitch for the right ones. Give it the kind of role you want, your recent titles, level, years, skills and where you would work, and it ranks the open jobs at up to three companies by how well their titles, level, skills and location match, with the reasons and the gaps for each. For one job it lists the must-haves, nice-to-haves, level signals and stated hiring process, plus a checklist for tailoring your resume to it, based on general recruiter practice.
You can also check your own profile text against the words recruiters search for. Paste your headline and About section, name the roles you want or one job, and it shows which typical title, skill and tool terms your text has and lacks, an example search string, and tips for writing the headline and About section as a career story. It never reads LinkedIn or any site: you paste the text.
The ranking is a rule-based match against the public postings, not a prediction of whether you will be hired. It never writes your resume: your assistant does that from the facts it returns.
It stores nothing about you: answers about jobs are cached for up to two hours, what you tell it about yourself is used only while it answers and is never cached or kept, and the only lasting record is a daily count of calls per tool.
What it can do
- Lists the open jobs at a named company from its public careers board
- Narrows jobs by keyword, location, remote work, department and posting date
- Shows only jobs that publish a salary when asked
- Compares openings at up to five companies in one answer
- Links every job to its application page
- Ranks open jobs for a profile by title, level, skills and location, with reasons and gaps
- Shows a job's must-haves, level signals and stated process, with a resume tailoring checklist
- Checks pasted profile text for the title, skill and tool words recruiters search, with an example search string
Tools
find_company_jobs, Find jobs at companies. Find open jobs at specific companies. Use it for questions like "jobs at Stripe", "is Notion hiring designers?", "remote backend jobs at Ramp and Linear" or "which Spotify jobs list a salary?". Pass company names as the user says them, or careers page links, up to 5 at a time. Covers companies whose careers pages run on Greenhouse, Lever, Ashby, Workable or SmartRecruiters; others come back as not_found. Optional filters: keyword, location, remote, department, postedSince, withSalary. Returns each job's title, locations, workplace type, salary when published, posting date, a short summary and the application link.rank_jobs_for_profile, Rank jobs for a profile. Rank the open jobs at up to 3 companies for a person's profile. Use it when the user wants to know which jobs suit them, for questions like "which of these jobs fit a senior backend engineer with 7 years of Go?". Pass companies and the profile the user gave: targetTitles, and where known recentTitles, level, yearsExperience, skills, locations and openToRemote. Every posting is scored on title, level, location and then, for the best ones, skills and years asked. Returns the jobs ranked with a score, a level-fit verdict (match, step_up, stretch, step_down), the reasons it fits and the gaps. A rule-based match to public postings, not a prediction of being hired. The profile is used only while the call runs.job_fit_brief, Brief on one job. Get the facts of one job and a checklist for tailoring a resume to it. Use it for questions like "what does the senior backend job at Stripe require?" or "how should I tailor my resume for this role?". Pass company and job (the applyUrl from find_company_jobs, or the title). Returns the posting's must-haves, nice-to-haves, level and the words that signal it, terms to mirror, the years asked, location, process and referral facts it states, and a short tailoring checklist based on general recruiter practice, which is no guarantee of a result. To compare the user's own profile text with a job's keywords, use profile_keyword_check.profile_keyword_check, Check profile keywords. Compare the user's own pasted profile text (headline, About section or resume) with the keywords recruiters search for a role or one job. Use it for "does my profile have the right keywords for backend jobs?", "what words am I missing for this job?" or "how do I make my LinkedIn headline findable?". Pass profileText (headline on the first line) and either targetTitles or company and job (the applyUrl or the title). Returns, per role, the title variants, skills, tools and seniority words present or missing, an example boolean search string and whether the text would match it, and short tips for writing the headline and About section as a career story. Fixed lists and the posting's own terms, no AI model; general recruiter practice, not a guarantee. It never reads LinkedIn or any site: the user pastes the text, which is used only while the call runs.
Add Company Jobs Finder in Pi
Pi does not speak MCP on its own. The pi-mcp-adapter extension (these steps match version 2.34.0) loads remote servers and keeps their schemas out of the prompt until you ask. Install it once, then restart Pi:
pi install npm:pi-mcp-adapterAdd the server to a shared MCP file. Use .mcp.json in the project if only that project needs Company Jobs Finder, or ~/.config/mcp/mcp.json if you want it in every project. Pi also reads ~/.agents/mcp.json. A Pi-only override lives in ~/.pi/agent/mcp.json or .pi/mcp.json and wins over the shared file.
{
"mcpServers": {
"company-jobs": {
"url": "https://jobs.openkrill.app/mcp"
}
}
}The url is enough. Do not set command: that field is for a local process. No API key header. After restart, run /mcp and confirm the server is listed. The first call connects lazily. Ask: What jobs is Stripe hiring for right now? The adapter should offer find_company_jobs, rank_jobs_for_profile, job_fit_brief, profile_keyword_check.
Leave directTools off unless you want those schemas in every turn. With a handful of tools the cost is small. The default proxy is the better fit: search for the tool, then call it. /mcp disable company-jobs writes a project override in .pi/mcp.json without editing the shared file. Run /reload after that change.
Pi will not find this server by scanning Cursor or Claude config unless you run /mcp setup and import it. Putting the url in .mcp.json is the direct path and the one this page tests against. The server is the same public endpoint the other clients use, with the same limits and the same privacy rules.
A call checked on 2026-10-01
Stripe publishes on a public Greenhouse board, which is one of the five boards this plugin reads. The request below was posted to https://jobs.openkrill.app/mcp as tools/call. HTTP 200. Source: the Company Jobs Finder server, read 2026-10-01.
{
"method": "tools/call",
"params": {
"name": "find_company_jobs",
"arguments": {
"companies": [
"Stripe"
],
"limit": 2
}
}
}- companies. 1 returned
- companies status. ok
Repeat the call yourself if you need a newer reading. Cached answers expire. A rate limit is not a result: wait and try again. Nothing in the call is a ranking, a filing, or advice.