Skip to content

AAK-MCP-CCTEMPLATES-CVE-2026-73222-001: claude-code-templates below 1.29.4 (--studio server: unauth 0.0.0.0:3444 → command-injection RCE)

Severity
HIGH
Category
Supply Chain (supply-chain)
Scanner
agent_audit_kit/scanners/mcp_cve_pins_2026_07.py
Auto-fixable
No.
SARIF name
ClaudeCodeTemplatesStudioCommandInjection

What it detects

Claude Code Templates (claude-code-templates), a CLI for configuring and monitoring Claude Code, before 1.29.4 launches the Claude Code Studio server (cli-tool/src/sandbox-server.js) via the --studio option bound to all interfaces on port 3444, with cross-origin requests permitted and no authentication. POST /api/execute passes the request-body prompt field to executeLocalTask() and POST /api/install-agent passes the agentName field to a child process; the same unsafe agent path is reachable from /api/execute through checkAndInstallAgent(). These attacker-controlled values reach child_process.spawn() with shell execution enabled, so shell metacharacters are interpreted as OS commands — arbitrary command execution with the developer's privileges. An attacker who can reach the port directly, or who lures a developer running Studio to a malicious website, compromises source code, credentials, and local data (CVE-2026-73222, CVSS 8.8). Fixed in 1.29.4 (the version before it is 1.29.2; no 1.29.3 was published, so treat < 1.29.4 and unpinned as exposed).

How to fix

Upgrade claude-code-templates to >= 1.29.4 and pin it. Do not run --studio on an untrusted network; bind the Studio server to loopback, require authentication, and reject cross-origin requests. Never pass request-body fields (prompt, agentName) into a shell.

To confirm the finding is gone, report only this rule:

agent-audit-kit scan . --rules AAK-MCP-CCTEMPLATES-CVE-2026-73222-001

Limitations

Reads claude-code-templates in requirements*.txt at any depth and, at the project root only, in pyproject.toml, Pipfile, package.json, .mcp.json, mcp.json, claude_desktop_config.json and the Pipfile.lock, poetry.lock, uv.lock, package-lock.json, pnpm-lock.yaml and yarn.lock lockfiles, taking the lowest version a range allows and reporting a reference that states no version. The dependency is not read when it is declared in a sub-directory manifest (packages/app/package.json), in .cursor/mcp.json or .vscode/mcp.json, or in a lockfile format the reader does not parse (a Yarn 2+ yarn.lock). A file over 2 MB, or a lockfile over 32 MiB, is not read. A declaration nested more than 64 levels deep in package.json, a TOML manifest or an MCP config (an MCP server more than 62) is still judged, and reported without a line; a lockfile entry keeps its line at any depth.

References


All rules. This page is generated from the rule registry when the documentation is built; to change it, change the rule in agent_audit_kit/rules/builtin.py. To give the rule a hand-written write-up, add docs/rules/AAK-MCP-CCTEMPLATES-CVE-2026-73222-001.md: the page then shows that text, with the registry reference appended under it.