skills/mukul975/anthropic-cybersecurity-skills/performing-paste-site-monitoring-for-credentials/Gen Agent Trust Hub
performing-paste-site-monitoring-for-credentials
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill retrieves data from well-known external platforms to aggregate threat intelligence. It fetches recent public pastes using the Pastebin Scraping API (scrape.pastebin.com) and performs code searches via the GitHub API (api.github.com).
- [DATA_EXFILTRATION]: The agent script transmits user-provided email addresses to an external service to identify security breaches. Specifically, it sends account identifiers to the Have I Been Pwned API (haveibeenpwned.com), which is a standard operation for this type of monitoring tool.
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection by processing untrusted data from external sources. 1. Ingestion points: The skill aggregates raw text content from public paste sites and GitHub Gists (SKILL.md, scripts/agent.py). 2. Boundary markers: None identified; the skill directly processes ingested text for pattern matching. 3. Capability inventory: The skill possesses the capability to perform network GET requests and write analysis results to the local file system. 4. Sanitization: Ingested content is processed using regular expressions but is not escaped or sanitized before being analyzed or saved to JSON alert files.
Audit Metadata