@4da/mcp-server
Summary: This MCP server gives AI coding agents dependency upgrade intelligence, vulnerability scanning, and persistent memory about a project's tech stack and decisions — all read from local lockfiles with zero config.
vulnerability_scan— Scans dependencies for known CVEs via OSV.dev across npm, Rust, Python, and Go.dependency_health— Reports version freshness, deprecation status, and CVE counts per dependency.upgrade_planner— Produces a prioritized upgrade plan (CVE severity, deprecation, version distance), separating quick wins from breaking changes.what_should_i_know— Pre-task briefing covering advisories, active decisions, signals, and ecosystem updates.ecosystem_pulse— Surfaces live Hacker News discussions filtered to the user's tech stack.get_context— Returns what 4DA knows about the user: role, tech stack, interests, and learned affinities.decision_memory— Records, lists, updates, or supersedes architectural and tech decisions (writes data).check_decision_alignment— Checks whether a proposed technology or pattern aligns with recorded decisions.agent_memory— Cross-agent persistent memory so discoveries, decisions, and warnings carry over between sessions (writes data).All read-only tools are annotated as such;
decision_memoryandagent_memoryare the two write-capable tools.
Allows fetching ecosystem news from Hacker News using Algolia's search API, filtered by the project's detected tech stack.
Queries the npm registry for package version freshness, deprecation status, and weekly downloads to assess dependency health and upgrade plans.
Queries the PyPI JSON API for Python package metadata, including version information and license normalization.
@4da/mcp-server has moved
The 4DA MCP server now lives in its own repository, with its full history:
https://github.com/4DA-Systems/4da-mcp-server
Install it as before; the package name is unchanged:
claude mcp add 4da -- npx @4da/mcp-serverAs a Claude Code plugin:
claude plugin marketplace add 4DA-Systems/4da-mcp-server
claude plugin install 4da@4daThis folder stays only so that older links keep resolving. Issues, releases and contributions go to the new repository.
The server reads this app's database. The schema it relies on is published
from here as src-tauri/contract/app-schema.sql,
generated from the migrations; CI checks every app schema change against the
server's queries before it merges.
Available Tools
9 toolsagent_memoryA
Cross-agent persistent memory: what one agent learns, any agent can recall. Call to store a discovery, decision, or warning, or to recall prior context before starting work.
check_decision_alignmentARead-only
Check whether a technology or pattern aligns with the developer's recorded decisions. Call BEFORE suggesting a major tech change, new library, or architecture shift.
decision_memoryA
Record, list, update, or supersede the developer's architectural and tech decisions. Call when the user makes, changes, or asks about a settled decision or convention.
dependency_healthARead-only
Dependency version freshness, deprecation, and CVE counts across npm/Rust/Python/Go. Call when the user asks whether their dependencies are outdated, stale, or need updating.
ecosystem_pulseARead-only
Live Hacker News discussions filtered to the user's tech stack. Call when the user asks what is new or trending in their ecosystem.
get_contextARead-only
What 4DA knows about the user: role, tech stack, interests, and learned affinities. Call FIRST when you need to know what the user works on before answering or recommending.
upgrade_plannerARead-only
Prioritized upgrade plan (CVE severity, deprecation, version distance), quick wins vs breaking changes. Call when the user asks what to upgrade, or after dependency_health surfaces problems.
vulnerability_scanARead-only
Scan dependencies for known CVEs via OSV.dev across npm/Rust/Python/Go, zero config. Call when the user asks about security, vulnerabilities, or CVEs, or before you recommend a dependency.
what_should_i_knowARead-only
Pre-task briefing: advisories, active decisions, signals, and ecosystem updates for a task. Call BEFORE starting any non-trivial task to get caught up first.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 9 tools
Some tools are clearly distinct (vulnerability_scan vs dependency_health vs upgrade_planner), but get_context, what_should_i_know, and agent_memory all deal with context/memory and could cause misselection. check_decision_alignment overlaps with decision_memory's 'check' functionality.
Tool names follow a consistent snake_case pattern, mixing noun_verb and verb_noun forms (vulnerability_scan, dependency_health, upgrade_planner) with a few verb-first names (get_context, check_decision_alignment). The style is consistent, but verb placement varies slightly.
9 tools is appropriate for a developer-assistant server covering security, dependency, and memory features. Each tool seems purposeful and the count is within the ideal 3-15 range.
The security/dependency workflow is well covered (scan, health, plan), and memory tools cover both recording and recalling. Missing actions like clearing memory or comparing dependency versions are minor gaps that agents could work around.
Maintenance
Related MCP Connectors
Package intelligence for AI agents across npm, PyPI, crates.io and deps.dev. No API keys.
Package intelligence for AI agents across npm, PyPI, crates.io and deps.dev. No API keys.
Trust-check any dependency for agents: OpenSSF Scorecard, licenses, CVEs, deps. 7 ecosystems.
CVE lookups (NVD) and dependency-manifest audits (OSV) for AI agents. No API keys.
Related MCP Servers
- AlicenseAqualityDmaintenanceAI-powered dependency vulnerability and breaking change analyzer that scans dependencies, identifies vulnerabilities via OSV.dev, and uses AI to assess real impact and suggest fixes.3Apache 2.0
- AlicenseNot gradedqualityFmaintenanceOpen source intelligence for AI agents — GitHub project-health scoring, package dependency-risk analysis, trending repositories, license checks, and side-by-side package comparison.MIT
- AlicenseAqualityDmaintenanceDependency security & health auditing for AI agents with no account or API key required.22MIT
- AlicenseAqualityDmaintenancePackage intelligence for AI coding agents that checks npm and PyPI package health, deprecation, vulnerabilities, bundle size, and compares alternatives.5MIT