$ loading_
Wrap Claude Code CLI as MCP tools for headless coding sessions and automation.
This MCP tool is an open-source MIT-licensed project with no declared API keys or remote endpoints in the provided material. However, it wraps Claude Code CLI for headless invocation, implying local code execution and corresponding data access exposure, so the overall posture is caution; source availability helps, but low adoption and unknown maintenance still warrant isolated use.
The material explicitly states that no keys or environment variables are required, and no API key, token, or other sensitive credential is mentioned; based on the provided facts, credential exposure risk appears low.
The material lists no remote host endpoints, and the README does not declare any third-party destination for data transfer; from the provided facts, no explicit network egress target is identified. However, because it wraps Claude Code CLI, the material does not clarify whether the underlying CLI may independently use network access at runtime.
The system flags this tool as executes-code, and the description says it wraps Claude Code CLI as tools and allows headless session invocation. That implies local process spawning and execution of corresponding operations. This is a common but high-impact MCP capability and should be run in a controlled environment.
The material does not specify what files, directories, or resources it can read or write. Since it operates via the local Claude Code CLI, it will typically inherit access to the local workspace or other user-authorized data available to that CLI invocation. There is no evidence of permissions beyond its stated function, but the access boundary is not clearly documented.
Positive factors include open-source availability, auditability, and an MIT license. However, it comes from a third-party registry, shows only 0 GitHub stars, has unknown maintenance status, and lacks README detail, so evidence of maturity and trust is limited. No clear red flags such as closed-source exfiltration or overtly malicious behavior are present, making caution more appropriate than high risk.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "claude-cli-mcp" yet — see the docs or source repo.
Use claude-cli-mcp to start a headless Claude Code session, scan TODO comments in the current repository, implement them, and return a patch with change notes.
Returns completed code changes, patch content, and a brief implementation summary.
Invoke Claude Code CLI through claude-cli-mcp, analyze core modules in the src directory, add unit tests for each module, and list uncovered edge cases.
Outputs new test files, test code suggestions, and a coverage summary.
Connect claude-cli-mcp to an MCP client workflow: when a requirement document arrives, have Claude Code generate an initial scaffold, README, and task checklist.
Produces a usable project starter, documentation, and a follow-up development task list.
Enable Claude Code to perform coding tasks through the OpenAI Codex CLI.
Query Claude Code transcript analytics for cost, safety, audit, and efficiency insights.
Connect Claude models to IDEs and agents for free coding assistance.
Make project docs instantly accessible so Claude Code understands architecture and conventions.
Manage Claude Code sessions with hot restart and context compaction commands.
Control Claude Code sessions in tmux with fine-grained orchestration and efficient extraction.