Amazon Web Services has published security advisory 2026-121-AWS documenting an argument injection vulnerability, CVE-2026-97662, in its open-source security-agent-mcp-server. The tool implements the emerging Model Context Protocol (MCP) standard, enabling autonomous AI code-review agents and IDE extensions to query security diffs. Flawed parameter parsing in the repository diff scan handler enables an attacker to supply leading hyphens that are treated as command flags by underlying version control binaries, breaking directory confinement.
Argument Injection Breakdown (CWE-88)
The Model Context Protocol establishes standardized JSON-RPC communication between LLM client applications and backend tools. In security-agent-mcp-server, the diff tool allows an LLM agent to inspect changes between two Git revisions:
{
"method": "tools/call",
"params": {
"name": "diff",
"arguments": {
"base": "main",
"target": "--output=/path/to/target/file"
}
}
}
Because the server passed the user-supplied target parameter directly into the git diff invocation without preceding positional argument delimiters (such as --), Git interpreted flags such as --output as command options. An attacker manipulating the target branch reference in a pull request could force the Git process to overwrite, create, or truncate critical files within the host environment.
Vulnerable Software Versions & Remediation Matrix
| Package / Module Name | Vulnerable Versions | Fixed / Remediated Release | Remediation Action Required |
|---|---|---|---|
| security-agent-mcp-server (PyPI) | Versions < 0.2.0 | 0.2.0 | Upgrade package via pip or poetry |
| awslabs/mcp GitHub Repository | Commits prior to tag v0.2.0 | v0.2.0 | Git pull and rebuild local virtualenv |
Step-by-Step Remediation Playbook & Patch Commands
AWS recommends immediate migration to the patched release:
- Upgrade Package via Pip / Poetry:
# Upgrade package via pip in active environment pip install --upgrade "security-agent-mcp-server>=0.2.0" # Alternatively, update poetry dependency poetry update security-agent-mcp-server # Verify installed release version pip show security-agent-mcp-server | grep Version - Audit CI/CD Runner Workspaces:
# Audit active MCP server configuration in Claude Desktop or IDE configuration cat ~/.config/Claude/claude_desktop_config.json | grep -i "security-agent-mcp-server" - Enforce Argument Sanitization Wrapper:
# Defensive code pattern to prevent argument injection in subprocess calls import re def sanitize_git_ref(ref: str) -> str: if ref.startswith("-") or not re.match(r"^[a-zA-Z0-9._/-]+$", ref): raise ValueError(f"Illegal Git reference: {ref}") return ref
Threat Hunting: Inspect Git Command Invocations
Audit developer workstation process telemetry for Git commands launched with unexpected output redirects:
# Query Linux auditd logs for git diff commands with file output arguments
ausearch -c git -m EXECVE | grep -E "diff.*--output"



