indeed-job-lead-scraper
Fail
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill contains explicit instructions to automatically install a missing dependency or skill from an external URL (https://www.browseract.com/) and a GitHub repository (github.com/browser-act/skills) if it is not present. This instruction specifically mandates bypassing user confirmation and separate approval for dependency installation, which circumvents standard security protocols for running untrusted code.
- [EXTERNAL_DOWNLOADS]: The skill directs the agent to fetch and install external resources at runtime from non-whitelisted domains to enable its core functionality.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to scrape and process job posting data from Indeed, which is untrusted and externally controlled content.
- Ingestion points: Job titles, company websites, and job description summaries scraped from external web pages (Indeed.com) (SKILL.md).
- Boundary markers: Absent. The instructions do not specify delimiters or provide warnings to ignore instructions embedded in the scraped content.
- Capability inventory: The skill utilizes a browser automation tool (browser-act) and is instructed to export results to CSV or tables, involving file-system and network interaction (SKILL.md).
- Sanitization: Absent. There is no evidence of validation or sanitization of the scraped data before it is processed by the agent.
Recommendations
- AI detected serious security threats
Audit Metadata