Library
85,340 resources from 11 public sources. 54 checked by hand; the rest are unreviewed.
Results
Looks like a goal we have a starting path for: I want to connect AI to my tools
-
imp
Agents & frameworksRun large AI chat models privately on your own NVIDIA RTX 50 or RTX PRO graphics card. On one RTX 5090, Qwen3.8-27B writes about 70 words per second, 1.4x faster than vLLM. Even a model almost four times larger than the card's memory runs. Works with apps made for ChatGPT or Claude. Starts with one docker command.
git clone https://github.com/kekzl/imp.git -
cordum
MCP serversThe action firewall for AI agents. Enforce policy and human approval before risky tool calls, shell commands, workflows, and production changes, with auditable evidence.
git clone https://github.com/cordum-io/cordum.git -
gateway
MCP serversMCP gateway with runtime security policy, tool-call-level control, and audit of agent actions.
-
mcpulse
MCP serversAnalytics for MCP servers. Query your tool calls, first-call success, retries and schema cost.
-
protect-mcp
MCP serversFail-closed Cedar policy gate + Ed25519 signed receipts for agent tool calls. Denies on any error.
protect-mcp@0.32.4 -
Hatch
MCP serversHosting for AI agents: publish a live website in one tool call, ephemeral or forever.
-
agent-runtime-guard
MCP serversAgent 运行时全栈拦截: allow/deny tool calls, Ed25519 receipts, blocks RCE, theft, SSRF, injection. stdio.
agent-runtime-guard@1.0.0 -
agent-observability
MCP serversTrace AI agent execution: every tool call, every error, every dollar. Open source, local-first.
agent-obs@1.0.7 -
mcpcut
MCP serversJournals every MCP tool call with secrets redacted; with a policy, holds risky calls for approval
mcpcut@0.4.2 -
Frugal
MCP serversThe open routing layer for AI tools: tool calls routed per policy, decision on every result.
frugal-mcp@0.2.0 -
mcp-audit
MCP serversTransparent audit proxy for MCP servers: logs every tool call, flags poisoning and rug pulls.
mcp-audit-proxy@0.1.0 -
MCP Safety Warden
MCP serversMCP proxy adding security scanning, behavioral profiling, risk gating, and safe tool call execution.
mcpsafetywarden@1.7.6 -
governed-mcp-gateway
MCP serversHTTP MCP gateway: principal on tools/call + SSE, allowlists, vault rotate, CHP gates.
@cubiczan/governed-mcp-gateway@0.1.0 -
paywall-mcp
MCP serversLightning-paywall sidecar: wraps any stdio MCP server, requires paid invoices before tool calls.
paywall-mcp@0.1.2 -
quill
MCP serversPre-execution gate for AI-agent tool calls. Touch ID approval. Tamper-evident audit log.
quillx@0.2.0a5 -
MaskFlow PII proxy
MCP serversMask PII in outbound MCP tool-call arguments, unmask the results. Indian identifiers included.
maskflow-mcp@0.1.2 -
MOVA MCP Connector
MCP serversThin MOVA MCP proxy for bindings, transport resolution, and upstream tools/call execution.
ghcr.io/mova-compact/mova-mcp-connector:0.1.0 -
agent-guardrail
MCP serversDeterministic policy firewall for AI agent tool calls - YAML rules, not a fuzzy risk score.
guardrail-mcp@1.0.2 -
SafeNode MCP Gateway
MCP serversPolicy proxy for MCP tool calls. A denied call never reaches the downstream server.
safenode-mcp-gateway@0.1.2 -
protect-mcp
MCP serversSigned receipts and Cedar policies for AI agent tool calls. Claude Code hooks, MCP gateway.
protect-mcp@0.5.3 -
Derbent
MCP serversGate for coding agents' MCP and built-in tool calls: rules, approvals and hash-chained receipts
-
Deus Witness
MCP serversMCP proxy: holds risky tool calls for passkey approval; signed receipts you can check offline.
deus-witness@0.2.77 -
GenPark Levenshtein Fuzzy Tool Matcher
MCP serversFast Levenshtein distance and fuzzy similarity matcher autocorrecting hallucinated LLM tool calls
https://github.com/Alpha-Park/genpark-levenshtein-fuzzy-tool-matcher-skill/releases/download/v1.0.1/genpark-levenshtein-fuzzy-tool-matcher-skill-v1.0.1.mcpb -
Provared MCP executor
MCP serversChecks each tool call against a passkey-signed permission, then writes a countersigned receipt.
provared-mcp-executor@0.4.2
Indexed entries aren't reviewed or tested by InnerVault. Details come straight from each source. Anything that can reach your accounts, files or browser can do damage, so read the source before you install it.