cold-outreach-sequence
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill instructions are focused on guiding the agent through a specific sales workflow. No instructions were found that attempt to bypass AI safety guidelines, extract internal prompts, or override system-level restrictions.
- [DATA_EXFILTRATION]: The skill handles prospect details (names, roles, companies) for the purpose of generating outreach sequences. It uses web searches to gather public information. No patterns were found indicating the unauthorized collection or external transmission of sensitive local files, credentials, or environment variables.
- [COMMAND_EXECUTION]: The skill references the use of
web_searchtools to find news and LinkedIn signals. These are platform-integrated tools used for their intended functional purpose. No arbitrary shell command execution or system-level manipulation was observed. - [REMOTE_CODE_EXECUTION]: There are no patterns involving the download or execution of remote scripts, binaries, or packages.
- [NO_CODE]: The skill is implemented entirely via markdown instructions and logic definitions. It does not include any executable code files (such as Python or JavaScript), which significantly reduces the technical attack surface.
Audit Metadata