Stores what AI agents learn — thoughts, decisions, session events and repo facts — in PostgreSQL with pgvector and serves it back through versioned MCP tools with hybrid vector-plus-full-text retrieval and budgeted context packs. Enables multiple agents to share one scoped, role-isolated long-term memory, with deliberate promotion of distilled candidates into durable memories.

Enables Claude to brief self-hosted GrokBot bots through their webhook routines and then collect their replies in the same thread, with tools for waiting on replies, checking the inbox, and reading conversations. It includes a self-hosted dashboard for registering bots, managing encrypted webhook keys and tokens, and connecting Claude apps over OAuth or API tokens.

Enables AI agents and LLM platforms to create and manage live 3D isometric sessions, upsert agents, update progress and status, trigger speech bubbles, and remove agents through stateful MCP tools over STDIO or HTTP.

Enables Claude to search a CRM account's contacts and deals, add notes, and create follow-up tasks, exposing read-only and additive-write tools over a bearer-token-protected Streamable HTTP endpoint. No tool can delete or edit existing records.

Enables MCP hosts that require a local stdio process to connect to the hosted public demo, forwarding tool discovery and calls for fictional scheduling scenarios, product information, draft-sharing preparation, and optional feedback.

Enables phrase-aware wrapping of Korean headlines into up to three lines at a specified approximate display width, using punctuation and Korean connective endings to avoid awkward word splits. Runs locally as a deterministic, read-only tool that preserves the original wording and returns both line arrays and text.

Enables agents to scan course structures, play videos at real 1x speed, track document task completion, capture quiz screenshots for vision-model reading, select and submit answers, and verify scores through MCP tools.

Enables host assistants to discover and read the toolkit’s research methods over a local stdio MCP connection, supporting research planning, literature work, scientific reasoning, writing, figures, and academic document workflows. It does not itself search scholarly databases, generate research results, compile documents, or call models.

This MCP server lets an agent run local binary inspection on ELF, Mach-O, PE, APK, IPA, ASAR, WebAssembly, and firmware targets — pulling symbols, sections, classified strings, disassembly, and mitigation posture, and flagging dangerous imports and CWE patterns. It exposes eight stdio JSON-RPC tools (with no external API keys) that feed a multi-agent swarm for attack-surface mapping and exploit-chain synthesis.

Enables agents to discover, search, compare, and inspect tested MCP tools by problem or keyword, with certification verdicts, scorecard summaries, certified scope, and explicit not-tested details.

Enables running NXOpen journals and Python scripts inside an already-open Siemens NX 2512 session from chat, with tools for pinging/checking the live bridge, creating and opening parts, and listing modules. It also adds a left-sidebar NX panel showing the bridge endpoint, available tools, and copy-ready commands.

Enables agents to analyze OpenAPI contracts for deterministic integration risks, plan auth, sync, action, and webhook candidates, and diff contract versions for likely breaking changes through read-only tools.

A gateway MCP server that lets agents reach any upstream MCP server through one endpoint while enforcing policy rules, holding risky tool calls for human approval, containing prompt-injected sessions, and logging every call.

Enables Claude and Cursor to search a team's synced Slack, GitHub and Notion content through permission-aware hybrid search with citations, retrieve a source document with its review and comment thread, and query time-aware knowledge graph facts when the graph is enabled. Every tool call runs with the signed-in user's permissions, so the assistant sees exactly what that user can see.

Enables AI agents to read messages from user-approved WhatsApp chats via a blocking long-poll tool and send signed replies to chats where the user has explicitly allowed responses. Permissions, watch/reply switches, and a live feed of everything the agent sees and sends are managed through a local web UI.

Enables AI hosts to retain evidence, check explicit constraints, plan bounded work, and record outcomes through local-first MCP operations. Optional reconstruction tools also support experimental image-to-3D workflows with local model providers and Blender.

Enables evidence-first AWS architecture work through deterministic design calculations, live official AWS knowledge, integration catalog/configuration, and optional authenticated local read-only AWS tooling for requirements, comparisons, diagrams, ADRs, and operational planning.

An MCP proxy that sits in front of bank tool servers, letting AI agents make read-only calls like balances and transactions while blocking payments, writes, renamed tools and argument smuggling. Each denial produces an Ed25519-signed receipt that can be verified offline.

Keeps large tool outputs out of an agent's context window by spilling them to disk and handing back a short pointer instead. Lets the agent pull only what it needs — exact numbered line ranges read directly from disk, or verbatim relevant lines returned by a small model answering a question about the stored output, plus tools to open, list, and close documents.

Enables agents to probe URLs and classify bot-wall responses with vendor and signal evidence, perform offline classification, scan for exposed key-like material, and run owner-side access audits through stdio tools.

Experimental Windows-first MCP server for Adobe Photoshop, with guarded brush-based painting and direct document and layer tools through a local UXP bridge. Requires local Photoshop on Windows for execution; Linux supports MCP catalog introspection only.

Composable mathematics tools for agents

Enables an LLM to orchestrate multiple downstream MCP tools inside one isolated sandbox in a single turn, filtering intermediate results in memory and returning only the final output to cut context tokens.

Turns any API with an OpenAPI 3.0/3.1/3.2 spec into callable MCP tools: it generates a tool per supported operation, validates arguments, and forwards authenticated requests, served over stdio or local HTTP or embedded as a library. No hand-written tool per endpoint is required.