Lists all MCP servers configured in your AI client, with environment variables masked. Helps you see existing setups before adding new servers or find exact server names for testing or removal.
Analyze a project directory to identify its tech stack and get recommendations for MCP servers that integrate with it. Checks already installed servers to show what's missing.
Count AI-agent configuration in a directory: instruction files, hooks, MCP servers, skills, subagents. Use before working in an unfamiliar repo or reviewing PRs that change agent behavior.
Generate a standards-compliant SBOM for your AI agent setup by discovering agents and MCP servers, extracting dependencies, and outputting in CycloneDX or SPDX format.
Identify which monitored MCP servers depend on a given npm package. Use during incident response to assess blast radius and determine impact across your infrastructure.
Unload MCP server tools from the current session to free context. Servers remain configured and can be reloaded later. Idle servers auto-unload after 10 calls to other servers.
Recommends MCP servers by filtering with AgentForge Trust Scores based on a natural-language use case, using semantic search and trust criteria to find reliable servers.
Scan MCP servers for security vulnerabilities including prompt injection, tool poisoning, and credential exposure. Returns findings with severity levels.