linkedin-mcp-custom
Analyzes LinkedIn saved jobs with EROI scoring and writes analysis back to a knowledge base for automated reporting.
README
linkedin-mcp-analyzer
MCP server for automated LinkedIn saved jobs analysis with EROI scoring and KB write-back.
Architecture
LinkedIn (saved jobs) → Patchright browser → Scraper → EROI scorer → KB writer
6 dimensions: domain 35%, tech 25%, role 20%,
growth 10%, formal 5%, location 5%
→ agregovany_report.md + metadata_stacku.json + git commit
Setup
uv sync
linkedin-mcp --login # one-time LinkedIn auth
linkedin-mcp --status # verify session
Usage
# Start MCP server
linkedin-mcp
# Or via MCP client
uv run python -m linkedin_mcp_custom
MCP Tools
| Tool | Description |
|---|---|
get_saved_jobs |
List saved jobs from LinkedIn tracker |
get_job_details |
Scrape full posting for a job ID |
analyze_saved_jobs |
Full pipeline: scrape → EROI → KB write-back |
EROI Scoring
Dimensions
| Dimension | Weight | What it measures |
|---|---|---|
| Domain | 35% | Industrial automation vs adjacent vs noise |
| Tech | 25% | Skill overlap (content-aware match ratio × coverage) |
| Role | 20% | Engineering role vs "fake engineer" (service/sales) |
| Growth | 10% | Strategic employer (Siemens, Google…) vs growth vs other |
| Formal | 5% | Degree requirements with flexibility detection |
| Location | 5% | Remote/hybrid/CZ vs distant/office-only |
Thresholds
| Score | Verdict |
|---|---|
| ≥65 | SLEDOVAT |
| 50–64 | MEDIUM |
| 40–49 | HRANICNI |
| <40 | NESLEDOVAT |
Special patterns
- Fake engineer: title says "Engineer" but content is service/sales
- Positioning match: strong role match compensates for domain gap
- Degree flexibility: "equivalent practical experience" adds ~5%
- Electronics manufacturing: SMT/PCBA keywords cap domain score
Tests
uv run python tests/test_eroi_regression.py # 6 regression tests
uv run python tests/test_kb_writer.py # 4 KB writer tests
Phases
| Phase | What | Tag |
|---|---|---|
| 0 | Project scaffold | v0.1.0 |
| 1 | Browser + auth (Patchright) | — |
| 2 | Scraping engine (LinkedInExtractor) | — |
| 3 | MCP tools (get_saved_jobs, analyze…) | v0.2.0 |
| 4 | EROI engine (6 scorers) | v0.3.0 |
| 5 | KB writer (report + metadata + git commit) | v0.4.0 |
| 6 | DevOps (ruff, pre-commit, README) | v0.5.0 |
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.