FRENES

Security · MCP server

Csl core: MCP server, setup and security

Csl core is an open-source MCP server that lets an AI agent write, formally verify and enforce machine-verifiable constraints for AI agents. Its repository has 27 stars and its last commit is from October 5, 2026 and it is licensed under Apache-2.0.

Actively maintained

What Csl core does

According to its repository: “Deterministic AI safety policy engine with Z3 formal verification: write, verify and enforce machine-verifiable constraints for AI agents.”

An MCP server gives an AI agent extra tools it calls during the conversation: the agent reads their description, decides to use them and gets the result back. Csl core is listed under Security, a category of 133 servers in the directory, where it ranks #60 by stars.

Topics declared by the repository: ai-governance, ai-safety, deterministic-ai, formal-verification, langchain, policy-as-code.

Installing and connecting Csl core

No package or remote server is declared in the official MCP registry: you install it from source, following the repository README.

Like any MCP server, you declare it in your client's configuration (Claude Desktop, Claude Code, Cursor, VS Code, Codex…): a name, the command that starts it or its URL, and any environment variables. The client then starts it and lists its tools. Project language: Python. Listed systems: macOS, Windows, Linux.

Security: what to check

An MCP server runs code with the rights you give it, and the model reads its tool descriptions: a malicious or compromised server can hijack the agent. Read the code, or pick a maintained and widely used server, before connecting it.

None of the risk signals we track (system access, secrets, missing licence, abandonment) applies to Csl core. That does not replace reading the code.

Good practice: grant only the rights needed (least privilege), require human approval for irreversible actions, and run the AI agent security check.

Project activity and maintenance

Created on February 6, 2026, the Chimera-Protocol/csl-core repository has 27 stars and 13 forks, with 14 open issues. It has more stars than 43% of the servers in the directory. It received changes this month.

Stars measure attention, not quality: a recent, heavily promoted project can collect them fast, while a useful but niche server will have few.

World MonitorWorld Monitor : 89.4K89.4KSemgrepSemgrep : 17.2K17.2KIda pro MCPIda pro MCP : 13.1K13.1KGhidraMCP (LaurieWired)GhidraMCP (LaurieWired) : 11.5K11.5KRobin: AI Powered Dark Web ORobin: AI Powered Dark Web O : 7.7K7.7KJadx ai MCPJadx ai MCP : 2.9K2.9KPower BI Authoring MCP ServePower BI Authoring MCP Serve : 1.2K1.2KSafeDep Vet MCPSafeDep Vet MCP : 1.1K1.1KCsl coreCsl core : 2727
GitHub stars of the most followed servers in the category (log scale); Csl core in orange. Checked October 25, 2026.

Verdict

Csl core is a reasonable choice; check the points below before connecting it.

Strengths

  • Active project: recent commit
  • Permissive licence (Apache-2.0)

Limitations

  • Install from source

Frequently asked questions

How do I install Csl core?

No package is declared in the MCP registry: clone the Chimera-Protocol/csl-core repository, follow its README, then declare the start command in your client's MCP configuration.

Is Csl core free?

The code is published under the Apache-2.0 licence and can be used free of charge.

Is Csl core safe?

No risk signal found (system access, secrets, licence, abandonment). As with any MCP server, keep rights minimal and read the code before sensitive use.

Is it an official server?

Nothing says so: it is a project published by Chimera-Protocol on GitHub. Check who maintains it before giving it access.

What are the alternatives to Csl core?

In the same category: Semgrep, World Monitor, SafeDep Vet MCP, and Ida pro MCP. Semgrep and World Monitor score higher on our grid.

Related articles

Neutrality. No affiliate links. No vendor pays to be reviewed, rated or ranked. Reviews are directed by François Chollet, neutral and factual, and every data point is dated and sourced. How we score