linkedin-ads
Pass
Audited by Gen Agent Trust Hub on Jul 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions for the user to install the vendor's official MCP tool using
npx -y @coldiq/mcp@latest. This package is a legitimate resource from the author, used to provide the agent with specialized advertising intelligence capabilities. - [COMMAND_EXECUTION]: Instructions in the setup section involve executing shell commands to configure environment variables and run the vendor's package. This is a standard integration method for connecting the agent to the Cold-IQ API marketplace.
- [PROMPT_INJECTION]: The skill enables the agent to ingest and analyze external data from LinkedIn Ads and competitor research through tools like
mcp__coldiq__search_ads. This creates an attack surface for indirect prompt injection if external content contains malicious instructions disguised as data. - Ingestion points: The agent reads data from competitor ads and external research via tool outputs and provided reference URLs.
- Boundary markers: There are no explicit instructions or delimiters defined to separate untrusted external text from the agent's internal logic.
- Capability inventory: The skill has the capability to read local files and invoke external tools to build audiences and research competitors.
- Sanitization: The skill does not describe any specific filtering or sanitization of text retrieved from the LinkedIn Ads Library or other external sources.
Audit Metadata