Enables MCP clients such as Claude Desktop, Claude Code and Cursor to OCR local scanned PDFs and images in 100+ languages, returning the extracted text or a searchable PDF. It also lets agents check document status, list recent documents and delete them.

Local, read-only MCP tools for B2B outreach: French telemarketing consent validity (décret 2026-662), GDPR gap report for customer service outsourced outside the EU, cost per kept meeting by lead source, and least-cost capable model selection.

Enables AI assistants to read and write Banktivity .bank8 vaults through Core Data, covering accounts, transactions, splits, categories, payees, tags, templates, import rules, scheduled transactions, statements, and securities while keeping edits visible to CloudKit sync.

Provides trust and commerce checks for agent operators through HTTP x402 and a remote MCP endpoint, enabling clients to inspect x402 quotes, MCP handshakes, and A2A cards and to run paid verification and data tools.

Enables AI agents to generate studio-grade fashion photography and 4K motion-captured video, perform garment try-ons and background edits, and export creative assets to commerce, productivity, and social platforms.

Enables AI agents and foreign investors to search Korean corporate disclosures, retrieve filings and XBRL financials, and access English-translated governance and accounting terms with original Korean text preserved for citation.

Enables an AI assistant to search live neighborhood bulletin-board pins by coordinates, radius, bounding box, or category, and to fetch individual pins or share their page links. It also lets the assistant publish public pins such as events, lost pets, sales, and business notices, gated behind explicit user confirmation and optional paid placement plans.

Enables AI coding sessions across different coding tools to collaborate through end-to-end encrypted messaging, work handover, and change review, while the relay stores only ciphertext.

Enables searching USDA AMS Livestock Mandatory Price Reporting market news reports and reading dated price rows — direct slaughter cattle, boxed beef and pork cutouts, hogs and lamb — keeping low, high and weighted-average prices separate with each report's own basis fields. It also reports clearly when a requested date, section or commodity family has no AMS coverage instead of returning empty results.

Enables point and radius queries against live French public geodata, covering commune register details and population, nearby building footprints with attributes, CORINE land-cover class at a location, and recorded flood-risk zones. All lookups are proxied per request with no stored copies.

Provides AI agents with live access to Chile's financial regulator data: sanction resolutions, regulations (NCGs, circulares), and registers of supervised entities and Fintec financial service providers. Supports sanctions lookups, norm resolution by type and number, and entity lookup by RUT or name, with natural persons' data withheld.

Read-only checks for money that agents move. verify_receipt: schema, Ed25519 signature, mandate expiry, amount = quantity x unit_price, refund not above payment. audit_spend: duplicate charges, price escalation, failed-but-billed, runaway loops, missing refunds, model overpay, budget breaches - three separate amounts, never one. reserve_report: NAV with an evidence hash per line; refuses a backing

Enables agents to set up and manage named connections to a user's own machine, keeping credentials in private local files while requiring explicit permission review. Runs locally over stdio with no hosted endpoint and preserves existing manager state.

Act for a person on websites: sandboxed browser, phone approvals, saved logins, scheduled plans.

Open Plugin Market — Community plugin index for DeepSeek Harness (DSH). 照顾小开发者的插件索引、插件市场、开放插件市场

Create and refine Blender product scenes in DSH: editable recipes, model imports, real previews, saved revisions and PNG/MP4 delivery.

DSH permission-gate for the DeepSeek Harness 0.1.5 line: a single self-sufficient, deterministic-first, fail-closed gate covering P0 hard-deny -> P1 session grant -> P2 static rule (allow/deny) chain -> P3 optional LLM semantic classifier -> P4 ask, with

Free open-source OSINT and investigation toolkit for Claude Code: case management, 54 OSINT enrichment adapters, structured analytic techniques (ACH), chain-of-custody evidence capture, and threat-intel MCP servers. MIT.

LangSmith plugins for Claude Code

Spec-driven development and context engineering for Claude Code, Cursor, Codex, and GitHub Copilot — backed by project context in Git.

Agent skills for engineering and design

Persistent local memory for AI coding agents — Claude Code, Codex CLI, Cursor, any MCP client. Temporal knowledge graph, procedural memory, AST codebase ingest, cross-project analogy. LongMemEval R@5 95.1%, LoCoMo 0.607, BEAM 1M 0.448. 74 MCP tools, 9 IDEs, 100% local.

Save 94% on AI coding tokens. Index your codebase, agents search instead of reading files. Works with Claude Code, Codex, Copilot, Cursor, Gemini CLI. Local MCP server, free, open source.

The missing reliability layer for spreadsheet work in LLM agents. Read, write, diff, validate, and analyze .xlsx files end-to-end — with merged cells, formulas, named ranges, conditional formatting, pivots, and charts preserved.