pinterest-research
Pass
Audited by Gen Agent Trust Hub on Aug 10, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses
curlto interact withapi.crawlora.net. This is the core functionality of the skill (Pinterest research) and targets the vendor's own infrastructure (crawlora-org). These network operations are expected and do not involve sensitive data exfiltration. - [COMMAND_EXECUTION]: A helper script
scripts/crawlora.shis provided to wrapcurlcommands. The script uses standard shell practices (set -euo pipefail) and includes safety measures like--data-urlencodeto prevent command injection through query parameters. Usage is limited to intended API interactions. - [CREDENTIALS_UNSAFE]: The skill correctly instructs users to manage their API keys using environment variables (
CRAWLORA_API_KEY) rather than hardcoding them. It includes specific warnings against committing or hardcoding keys. - [DATA_EXFILTRATION]: No evidence of unauthorized data exfiltration was found. The skill only accesses public Pinterest data (profiles, pins, boards) through the authenticated Crawlora API.
Audit Metadata