Claude MarketplacesSubmit
nugehs avatar

Aiglare

nugehs/aiglare

Point this at a JavaScript or TypeScript repo and it scans for every place an LLM output reaches a user or triggers a side effect like a payment, email, or database write. The MCP server exposes three tools: ai_surface_audit returns a full governance report flagging missing confidence checks, fallbacks, validation, and human-in-loop controls; ai_surface_gate gives you a pass/fail CI verdict; list_providers shows the detection registry covering OpenAI, Anthropic, Vercel AI SDK, LangChain, and a dozen others. It's static analysis, not runtime validation. Treat it as a linter that finds AI surfaces with no guardrails so you know where to add them. The CI mode fails only on red side-effectful paths, the "model auto-triggers an irreversible action" case.

GitHub stars
1
tools
3
transport
STDIO
last updated
Jun 10, 2026
View on GitHub

Tools

3 tools

Public tool metadata for what this MCP can expose to an agent.

ai_surface_auditFull audit of a repo (`path`, optional `sinks`, `severity`) → the same structured report as `--json`

Full audit of a repo (`path`, optional `sinks`, `severity`) → the same structured report as `--json`

No parameter schema in public metadata yet.

ai_surface_gateCI-gate verdict for a repo: `passed` + count of blocking red side-effectful surfaces

CI-gate verdict for a repo: `passed` + count of blocking red side-effectful surfaces

No parameter schema in public metadata yet.

list_providersThe provider registry the scanner detects

The provider registry the scanner detects

No parameter schema in public metadata yet.

Sponsors

Promote your product hereReach 125k+ visitors monthlyPromote your product hereReach 125k+ visitors monthlyAgent CaffeineSponsoredClose your lid, come back to finished workSnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthlyWaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthlyknobsSponsoredSee your site on any phone, any day, in any language

Readme

aiglare

Lint your AI features for governance guardrails — where can the model do something you can't undo?

npm CI license: MIT node

Live site: nugehs.github.io/aiglare-web

aiglare demo

Point it at any JS/TS repo and it finds every place an LLM/AI output reaches a user or triggers a side-effect (payment, booking, email, database write) — then flags which of those have no confidence handling, no fallback, no output validation, and no human-in-the-loop.

View the full README on GitHub

Details

Repository
github.com/nugehs/aiglare
Category
AI & LLM Tools
Package
@nugehs/aiglare
Transport
STDIO
Registry
MCP registry, active

More from nugehs

Related AI & LLM Tools MCP servers

View all
1prompeteer avatarPrompeteerprompeteer/prompeteer-mcpContextual prompts and agent skills for 140+ AI platforms.1stars
2pvliesdonk avatarImage Generation MCP Serverpvliesdonk/image-generation-mcpMCP server for AI image generation via OpenAI, Stable Diffusion (SD WebUI), or placeholders.1stars
3rafavecchi avatarGrantdrafavecchi/grantdGrantd MCP server: let your AI agent act on a user's behalf across third-party APIs via OAuth.1stars
4renl avatarMeshimize Mcprenl/meshimize-mcpMeshimize: Q&A groups, messaging, group discovery, and agent-to-agent delegation1stars
5rumblingb avatarAgentpay Feed Mcprumblingb/agentpay-feed-mcpReal-time agentic newsfeed for MCP tools and capabilities1stars
6samoradc avatarTetradsamoradc/tetradQuadruple consensus MCP server for code validation using Codex, Gemini, and Qwen1stars