Manage and interact with MCP servers and tools via the mcporter CLI, supporting listing, configuring, authenticating, calling tools, and generating CLI/types.

This skill appears to be an instruction wrapper for the mcporter CLI and is otherwise coherent, but there are a few issues to check before installing: - Confirm the install source: SKILL.md points to a Homebrew formula in the 'pdxfinder/tap' tap (a third‑party tap). Inspect the tap/formula on GitHub to ensure it builds the expected mcporter binary and doesn't contain malicious build/install steps. Prefer official or well-known release channels when possible. - Inspect local config: mcporter uses '~/.config/mcporter/' for server configuration; review any files there (or the files the CLI will create) since they can contain server endpoints and authentication tokens. Treat those as sensitive. - Metadata mismatch: the registry entry omits required binaries and config paths that SKILL.md references. Ask the skill author/maintainer to update the registry metadata to declare the 'mcporter' binary requirement and the config path so you can evaluate permissions more clearly. - Test in a sandbox first: if you want to be cautious, install and run mcporter in an isolated environment (VM, container) and review network activity and created files before allowing it access to your regular account and secrets. If the author provides an official release URL, an audited Homebrew formula, or updated registry metadata that declares the config path and binaries, my confidence would improve and the assessment could move toward 'benign.'

Type: OpenClaw Skill Name: mcporter-skill Version: 1.0.0 The skill bundle provides a wrapper for the `mcporter` CLI tool, enabling an AI agent to manage MCP servers and tools. The installation instructions use a standard Homebrew formula (`pdxfinder/tap/mcporter`), and the provided commands (`list`, `auth --help`, `call`, `generate`, `config --help`) are all legitimate uses of the `mcporter` CLI. There is no evidence of intentional harmful behavior, data exfiltration, malicious execution, persistence, or prompt injection attempts against the agent to perform unauthorized actions. The skill's behavior is clearly aligned with its stated purpose.