ptlam-scraping-urls
Pass
Audited by Gen Agent Trust Hub on Aug 29, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
statto check file modification times for cache validation. The usage is restricted to target files within the resolved output directory and does not incorporate unvalidated user input directly into shell execution. - [DATA_EXPOSURE]: The skill reads and writes files within a dedicated workspace folder (
.ptlam-agent-plugin/ptlam-scraping-urls/). It includes explicit instructions to reject path traversal attempts (e.g.,..) and requires explicit user consent for writing outside the workspace. - [EXTERNAL_DOWNLOADS]: The skill performs batch scraping of URLs provided by the user. While this involves network operations, it is the primary stated purpose of the skill and uses the host platform's standard tools for retrieval.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external URLs. However, it treats the content as data to be stored in Markdown files rather than instructions to be executed. The
retrieval.mdinstructions explicitly mandate JSON for data exchange to prevent delimiter-based injection during result reporting.
Audit Metadata