agent_extensions

2 registry entries were observed exposing a tool named agent_extensions — none of them flagged as a privileged capability.

Spelled agent_extensions, agent-extensions in the wild — all one tool for the purposes of this page.

Every row is point-in-time scan output, dated below. Forge has an observed tool surface for 12,691 of 42,202 scannable entries (30%) — an entry that has not been scanned cannot appear here, so absence from this list is not evidence that an entry does not expose this tool.

  1. zn-gateMCPattestedtrust 85/100 · A
    v1.8.2
    ↓ 506/wk
    ★ 2

    Local security layer for coding agents (Claude Code, Codex CLI, Cursor, OpenCode): keeps secrets out of what the agent sends to the model, checks for prompt injection and guards the MCP servers you wrap.

    ClaudeCursorCopilotGemini
    useznMITtodaynpmGitHubobserved · static extraction
    prompt-injectionmcpmodel-context-protocolai-securityllm-guardrailsai-agents
  2. whats-inherited-mcpMCPcommunitytrust 60/100 · B
    v0.1.6

    What a checkout tells your agent to do: instruction files, hooks, declared MCP servers.

    ClaudeCursorCopilotGemini
    io.github.stcmainUnknown2mo agonpmGitHubobserved · static extraction
    mcp

All indexed toolsSearch the registry for agent_extensions