MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
AI agent budget management & anomaly detection | x402 USDC | JPYC ready
-
agent-guard-mcpgithub.com/dygogogo/agent-guard-mcpupdated this year
MCP Guard Server - Budget control, approval workflows and audit logging for AI agents (Claude Code, Cursor, ChatGPT)
MIT -
A local MCP server for AI-assisted YNAB budget understanding with guarded assignment previews.
npm MIT -
Spending guard for AI agents that pay x402 paywalls in USDC: budgets, payee allowlists, price-spike checks and human approval before anything is signed. MCP server. Arc + NEAR testnet.
npm MIT -
Agent proposes, human signs: a budget guard between AI agents and send21's MCP server. Crypto pay links and API for people and AI agents.
MIT -
Alexa+ MCP server: budget-aware shopping guard for buyers, inventory co-pilot for sellers
MIT -
Local stdio MCP for KIE images and video: live model choice, guarded approvals, shared budgets and private owner setup.
MIT -
Governed AI-ops for OPNsense + pfSense firewalls — gateway/rule/traffic RCA, guarded rule/alias writes, with audit/budget/undo (preview)
pypi MIT ★ 2 -
Governed AI-ops for Docker + Portainer container hosts — restart-loop/resource/bloat RCA, guarded writes, with audit/budget/undo (preview)
pypi MIT ★ 1 -
model-context-protocolgithub.com/christian-dussol-ai-native/model-context-protocolupdated this year
A governed MCP server exposing Kubernetes cost data from OpenCost to an AI agent, under least privilege, audit and a budget guard.
NOASSERTION -
A multi-model council you can call like one model. MCP sidecar around OpenFusion (unmodified): panel → judge → synthesizer, with a budget guard.
MIT -
spendwatchgithub.com/cognis-digital/spendwatchupdated this year
Multi-provider LLM usage, cost & rate-limit meter with budget guards — Anthropic/OpenAI/OpenRouter/local, TUI + MCP + CI exit codes. Zero-dependency.
NOASSERTION ★ 1 -
Identify the exact product (SKU) from a customer photo: image embeddings + vision-LLM re-ranking, FastAPI + MCP tool, multi-tenant guards, eval gate, budget breaker.
MIT -
Hardened Windows launcher for meta-ads-mcp: credentials in Windows Credential Manager, log masking, write guard for budgets/deletes, token check, bilingual setup wizard (en/de).
MIT -
Rust MCP server for generative media with avatars: InsightFace/ArcFace identity core, ONNX face swap, budget-guarded pipeline. Work in progress.
NOASSERTION -
Hardened MCP server for live web data: free direct fetching and local Chrome rendering, Bright Data only for bot walls. 34 tools, budgets, caching, SSRF guard, stdio + HTTP.
NOASSERTION -
tokenslimgithub.com/himanshudhawale/tokenslimupdated this year
Shrink the token cost of context you feed to LLMs — count tokens, estimate cost across models, and slim files (with a CI budget guard) before you send them.
MIT -
A local harness for Claude Code and Codex: budgeted test runs with a stop rule, a guard for the tests you already have, and a token tree for every subagent.
MIT -
NL→SQL agent over a lakehouse with five deterministic control rings: AST allowlist guard, cost budget estimated before execution, role-based masking and a tamper-evident audit chain. Served over MCP.
Apache-2.0 -
Open-source Kubernetes FinOps operator: cost per namespace, team, GPU and LLM token, schedules that scale idle environments and cloud databases to zero, budgets, chargeback and right-sizing with a rollback guard. Helm chart, dashboard, REST API, MCP server.
Apache-2.0 ★ 7 -
AIMG-AI-based_Media_Generatorgithub.com/MartinKhadjo/AIMG-AI-based_Media_Generatorupdated this year
Local-first, cloud-ready, model-agnostic AI media generation platform: runs 14B video models on a 12 GB laptop GPU, auto-integrates new models hardware-aware, directs clips and multi-scene commercials via LLM with human-in-the-loop approval, enforces quality gates and a server-side budget guard, fully orchestrable by AI agents via MCP.
-
Zero-dependency runtime control for production Python agents - stop loops, retry storms, and budget burn
pypi MIT -
⚡ Terminal HUD, hard budget breakers & prompt-cache leak alerts for AI coding agents (Claude Code, Cursor, Codex)
MIT ★ 32 -
Zero-dependency CLI and MCP server for LLM token cost estimation, context budget enforcement, and prompt secret scrubbing.
MIT -
mcp-guarded-toolsgithub.com/ridwanspace/mcp-guarded-toolsupdated this year
MCP server treating a large tool surface as a governance problem: tool-search discovery, request and token budgets, human-in-the-loop confirmation for writes, and an append-only audit log.
MIT -
Spend approval gate for AI agents — enforce budget limits, per-asset caps and human-in-the-loop thresholds before an agent moves money, with a signed audit hash.
npm MIT -
x402check-mcpactive
MCP stdio bridge to x402check — preflight_check, verify_payto, compare_tools before paying (ok|caution|avoid + history) — check before paying; is this API a scam/safe.
npm MIT -
llm-guard-gatewaygithub.com/ShreyAC962/llm-guard-gatewayupdated this year
Governance gateway for LLM and agent traffic: blocks prompt injection, redacts PII, enforces per-tenant token budgets, and serves repeat prompts from a semantic cache, cutting served p50 from 44.2 ms to 4.6 ms in the included benchmark. REST + MCP surfaces, offline-first local profile.
MIT