What tools does Agent Infra Mcp provide?
It provides 18 tools: 8 memory tools (store, batch store, list, get, update, delete, search, stats), 7 drift detection tools (create, list, run, history,
@lulzasaur9192
Basic information
18 MCP tools for AI agent infrastructure — persistent memory with FTS5, API drift detection with cron scheduling, and HMAC-signed trust receipts for audit trails
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"agent-infra": {
"command": "npx",
"args": [
"-y",
"@lulzasaur9192/agent-infra-mcp"
],
"env": {
"AGENT_MEMORY_API_KEY": "",
"AGENT_DRIFT_API_KEY": "",
"TRUST_RECEIPT_API_KEY": ""
}
}
}
}Store a key-value memory in a namespace. Supports optional metadata and TTL for auto-expiration. Use namespaces to organize memories by agent, project, or topic.
Store multiple memories at once (max 100). Each memory needs namespace, key, and value.
List all memories in a namespace. Supports pagination and sorting by creation or update time.
Get a specific memory by namespace and key.
Update an existing memory's value, metadata, or TTL.
Delete a specific memory or an entire namespace.
Full-text search across memories in a namespace using FTS5. Supports metadata filtering and sorting.
Get statistics about memory usage — total count, namespace breakdown, storage info.
Register a golden test case for API drift detection. Monitors an endpoint and alerts when output changes from expected. Supports exact match, JSON subset, contains, and status code matching. Optional cron schedule for automatic monitoring.
List all registered drift detection tests with their status and last run info.
Manually execute a drift test right now. Returns pass/fail, diff details, and response time.
Get the run history for a specific test — see when drift was detected, response times, and diffs.
Update a drift test's configuration — change endpoint, expected output, schedule, or match mode.
Delete a drift detection test and all its run history.
Get a dashboard summary — total tests, drift rate, average response times, and recent drift events.
Issue an HMAC-SHA256 signed trust receipt for an AI agent action. Creates a tamper-proof record of what an agent did, when, and with what inputs/outputs. Use this to build audit trails for agent workflows.
Verify the integrity of a trust receipt. Checks HMAC signature to confirm the receipt hasn't been tampered with. Can verify by ID (online) or by providing the full receipt data (offline).
List trust receipts with optional filters. Filter by agent, action type, or date range.
Agent Infra Mcp is an MCP server that provides 18 tools for AI agent infrastructure, including persistent memory, API drift detection, and HMAC‑signed trust receipts. It is designed for AI agents that need long‑term state, API monitoring, and tamper‑proof audit trails.
Install with npx @lulzasaur9192/agent-infra-mcp. Configure it in Claude Desktop or Cursor by adding a JSON entry under mcpServers with the command npx -y @lulzasaur9192/agent-infra-mcp and three environment variables: AGENT_MEMORY_API_KEY, AGENT_DRIFT_API_KEY, and TRUST_RECEIPT_API_KEY.
It provides 18 tools: 8 memory tools (store, batch store, list, get, update, delete, search, stats), 7 drift detection tools (create, list, run, history,
AON MCP connects AI agents to real, buyable products and services with live pricing and tracked links—enabling builders to earn developer revenue share from attributed clicks and conversions.
XDC AI is a remote MCP server available at https://xdcai.tech.

FLORA is a remote MCP server available at https://agents.flora.ai. The FLORA MCP lets users generate, organize, and iterate on visual assets — hero images, moodboards, lifestyle shots, promo graphics — directly from a co

Connect Atwork to Claude, ChatGPT, Cursor or any MCP client and work with your real data: ask what is overdue, total a sheet by any field, add tasks, update rows, post comments and draft pages. The assistant acts as you
Open-source Node.js FHIR MCP server with SMART Backend Services, metadata-aware search/CRUD tools, compact responses, FHIRPath filtering, safe pagination, audit events, and terminology lookup.
Comments