mcp-config-sync
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONPRIVILEGE_ESCALATION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions to execute various shell commands and scripts (e.g.,
agent-config-sync/scripts/sync.py,sync-tools.ps1, andsync-tools.sh) to perform discovery, planning, and synchronization of Model Context Protocol (MCP) server configurations. - [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to discover and ingest MCP server definitions from local configuration files. These files represent an attack surface where malicious instructions could be embedded in server names, arguments, or metadata.
- Ingestion points: The skill reads configuration data from paths such as
~/.claude/_shared-mcp.json,~/.claude/profiles/*.json, and%APPDATA%\Claude\claude_desktop_config.json. - Boundary markers: No specific delimiters or warnings to ignore embedded instructions are present in the provided scripts or instructions.
- Capability inventory: The skill has the ability to read and write files and execute shell commands via Python, PowerShell, and Bash.
- Sanitization: The provided scripts do not appear to sanitize or validate the content of the JSON configuration files before processing or displaying them to the user.
- [PRIVILEGE_ESCALATION]: The
scripts/sync-tools.ps1documentation includes a sample command that uses-ExecutionPolicy Bypass. This is a standard method to run unsigned scripts but technically bypasses local PowerShell security restrictions.
Audit Metadata