In-depth architectural comparison of the GroundTruth MCP and Mcpskills Server MCP servers. Compare execution transports, security boundaries, tool capabilities, quality scores, and ready-to-paste client installation snippets for Claude, Cursor, Windsurf, and VS Code.
At a Glance & Executive Verdict
GroundTruth MCP
Security · Local stdio
Quality: 52/100 (Good) | Auth: No auth required
Mcpskills Server
Security · Local stdio
Quality: 59/100 (Good) | Auth: API Key required
Verdict Summary: Choose GroundTruth MCP if you need specialized Security tools running via a local process. Choose Mcpskills Server if your workspace requires Security integration with local subprocess execution. Both servers can be configured concurrently in your client's mcpServers manifest.
Which MCP Server Should You Choose?
Choose GroundTruth MCP when:
You need dedicated capabilities in the Security domain.
You prefer local stdio subprocess transport architecture.
Your security boundary fits: No auth required (Free / Open Source).
You have access to required keys: GT_GITHUB_TOKEN.
Primary tools included: 14 MCP tools for documentation, audits, search, comparison, and migrations, 107+ audit patterns across 18 categories, Exact file-and-line findings with live fix guidance.
Self-hosted MCP for live docs, code audits, snippets, and best practices across 445+ libraries. 14 tools, 107 audit patterns with file:line precision, BM25-ranked snippets, lockfile-aware version pinning, npm/PyPI/crates.io/Go fallback. No API keys, no rate limits, Context7 alternative. npx -y @groundtruth-mcp/gt-mcp
Pre-install trust layer for MCP servers, AI skills, and npm packages. Scores any repo or package across 15 signals (incl. OSV/KEV/EPSS vulnerability intelligence) with safety scanning for prompt injection, credential theft, and supply-chain risk; the autogate tool returns a go/no-go install decision. Listed in the official MCP Registry as io.mcpskills/server. npm: @mcpskillsio/server. https://mcpskills.io
Tools & Capabilities Breakdown
GroundTruth MCP Tools (6)
14 MCP tools for documentation, audits, search, comparison, and migrations
107+ audit patterns across 18 categories
Exact file-and-line findings with live fix guidance
BM25-ranked, version-aware code snippets
Fallback package lookup across npm, PyPI, crates.io, and pkg.go.dev
Optional GitHub token for higher request limits
Mcpskills Server Tools (9)
Ready-to-Paste Client Configurations
Paste either (or both) of these JSON server blocks into your client config file (e.g. claude_desktop_config.json or ~/.cursor/mcp.json).
GroundTruth MCP is categorized under Security and uses a local stdio subprocess. In contrast, Mcpskills Server belongs to Security using local stdio subprocess. Select GroundTruth MCP when you need capabilities focused on security and Mcpskills Server when you require tools for security.
Score any AI skill, MCP server, or GitHub repo for trustworthiness. Returns a trust score (0-10) across 4 dimensions: Alive, Legit, Solid, Usable. Accepts: owner/repo, GitHub URL, npm package (npm:@scope/name or @scope/name), Smithery URL, or OpenClaw URL. AI skills get enhanced safety scanning. Set MCPSKILLS_API_KEY for full reports.
scan_safety
Run a focused safety scan on an AI skill or MCP server. Checks for prompt injection, shell execution, network exfiltration, credential theft, obfuscated payloads, public network binding (0.0.0.0), and risky npm lifecycle scripts (preinstall/install/postinstall). Accepts any input format (owner/repo, npm package, Smithery URL, etc.).
list_packages
Browse curated, pre-scored AI skill packages organized by use case. Each package contains vetted skills with trust scores. Available packages: Claude Power User, Full-Stack Vibe Coder, Data & Research, DevOps & Infrastructure, Content & Marketing.
get_badge
Get a trust badge URL for any repo or package. Returns a shields.io-style SVG badge showing the trust score and tier. Embed in READMEs. Badge auto-updates hourly.
watch_repo
Start monitoring a repo or package for trust score changes. Alerts when score changes significantly (±0.3 points or tier change). Requires a paid API key.
check_watched
Re-scan all watched repos and check for score changes. Returns any repos whose trust score changed significantly since last check.
batch_check
Check up to 5 repos or packages in one call. Returns a trust assessment for each. Requires a Developer Pro or Team API key. Accepts any mix of formats (owner/repo, npm packages, registry URLs).
auto_gate
Should I install this? Returns a simple go/no-go decision with reasoning. Accepts any format: owner/repo, npm package, Smithery URL, etc. Returns { proceed: true/false, reason: "..." }.
build_stack
Recommend a vetted stack of trusted tools for a described task. Describe what you're building (e.g., "Next.js app with auth, payments, and AI chat") and get back a curated list of the highest-scoring repos in each relevant category, pre-scored and ready to install. Returns tool names, trust scores, tiers, and install hints. Use this instead of guessing which tools to recommend — every suggestion is backed by live trust data.