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 CLI agents to fold source documents into a cited Markdown knowledge base using an ontology, exposing tools to list, read, search, lint, and write pages only after user approval.

Enables AI agents to read a user's SRCE Moodle account data, including grades, deadlines, calendar events, course files, forum posts, and messages, with optional private-message sending gated by phone approval. It can run as a Cloudflare Worker or self-hosted via Node or Docker.

Enables MCP-capable agent CLIs on macOS to reuse an installed, unmodified OpenAI Browser Runtime to drive a connected Chrome or Edge browser (or a host's in-app browser backend) through js, js_reset, and turn_ended tools. It stays a thin bridge that preserves the Runtime's approval and security boundaries, failing closed when the client cannot handle approval requests.

Enables any AI agent to work inside your own signed-in Chrome or Edge, with risky steps gated behind approvals the agent cannot skip, receipts for every action, routines that replay without the AI, and per-site packs that read a site's own data. Everything runs locally, so pages, cookies and passwords never leave your machine.

Audit trail and traceability for AI agents. See what every coding agent did, block risky calls, ask a human for approval, and prove it with signed trust checks.

Give a cloud AI agent (Arena.ai Agent Mode, Claude, Cursor, ...) controlled access to your own machine over HTTP/MCP: files, shell, Chrome and the macOS desktop, with an approval gate, trust windows and an audit log.

Android companion for DeepSeek Harness | chat, goals, approvals & notifications from your phone, over your LAN. Kotlin + Jetpack Compose.

Agentic SEO operating system as a Claude Code plugin: project Wiki, DataForSEO workflows, content briefs, technical SEO, and human-approval gates.

Your pocket agent. Local-first AI agents on iOS and Android — real workspaces, tool execution with approvals, and your choice of model (DSH · Claude Code · Codex · GLM).

Enables an AI agent to sign, verify, encrypt, decrypt, and manage OpenPGP keys while private key material stays in gpg-agent and passphrases go only to pinentry. Adds human approval for high-risk operations, typed-fingerprint confirmation for destructive ones, and audit logging that never records plaintext.

Human-in-the-loop (HITL) for AI agents: ask a person for approvals or answers by push notification. One command connects Claude Code, Codex, Cursor, Gemini CLI and more.

Enables deterministic request screening and, after native owner approval, fixed-target DeadBolt enforcement actions such as kill, deny, revoke, and evidence retrieval through local stdio MCP. Model analysis can advise but cannot approve actions.

Gives browser agents a reviewed map of 79 real sites — entry URLs, access verdicts, and role/accessible-name button maps for 104 screens — so an agent can find where things are and know whether it may act at all. Safety rules in code halt irreversible actions such as pay, order, submit and delete pending explicit human approval, while per-site memory recalls which routes worked last time.

Gives AI agents a durable, locally hosted temporal memory built from a markdown vault you own, exposing hybrid semantic/lexical/entity search over events, threads, transcripts, entity relationships with validity dates, and patterns with confidence decay. Writes are approval-gated, so agents can draft, commit, and sync new memories while keeping every index rebuildable and nothing leaving your machine.

Exposes tools that let an agent read a CV, search the Arbeitnow job board for Berlin AI/ML working student roles, score each vacancy against the CV, draft cover letters whose every claim cites CV facts, and queue them for human approval. The same tools can be driven from other MCP clients such as Claude Desktop, with scoring done deterministically in code and limits enforced outside the prompt.

Exposes agent-facing tools that let a personal AI agent submit each proposed action for an allow, require_approval, or deny verdict against OWASP PAA10 policy rules, plus tools to record provenance-tagged memories, log refusals, read status, and trigger a self-stop. Its decisions are advisory unless paired with the project's gating proxy or hook, which enforce them.

Human-in-the-loop (HITL) for AI agents: ask a person for approvals or answers by push notification. One command connects Claude Code, Codex, Cursor, Gemini CLI and more.

Open-source, self-hostable platform that turns agents into co-workers. Start with Software Factory: delegate GitHub issues and review pull requests, with model, approval, and cost controls in your infrastructure.

Human-in-the-loop (HITL) for AI agents: ask a person for approvals or answers by push notification. One command connects Claude Code, Codex, Cursor, Gemini CLI and more.

Connect AI agents to your own Chrome with scoped access, approvals, and verified task results.

Sovereign AI agents for business: answers from company data, each person's permissions, approvals.

Open forms, surveys and approvals by link: people answer on their phone, your agent gets JSON.

Human approvals, autonomy limits and a chained audit log for the AI agents of service firms.