offsite-opportunity-finder
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Connects to the Promptwatch MCP server at
https://server.promptwatch.com/mcpto retrieve citation data and model coverage statistics. - [COMMAND_EXECUTION]: Instructs the user to run
setup-promptwatch-mcpto configure the environment, which is a utility provided by the vendor for initialization. - [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it processes and summarizes content from third-party domains and pages.
- Ingestion points: Data is retrieved from external URLs using tools like
get_citationsandget_citation_top_pagesduring the workflow. - Boundary markers: There are no specific instructions or delimiters mentioned to prevent the agent from being influenced by malicious instructions embedded in the analyzed third-party pages.
- Capability inventory: The skill's tools focus on analysis and reporting (listing projects, retrieving citations, and identifying competitors) and do not appear to expose dangerous system capabilities.
- Sanitization: The instructions do not specify any validation or sanitization steps for the data retrieved from external sources before it is interpreted by the agent.
Audit Metadata