soia-safe-track-vulnerability-intel

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches vulnerability data from official industry and government repositories, including NIST's National Vulnerability Database (NVD), CISA's Known Exploited Vulnerabilities (KEV) catalog, and GitHub Security Advisories.
  • [COMMAND_EXECUTION]: Executes a local processing script (scripts/collect_vulnerabilities.py) to aggregate and normalize data. The script relies entirely on Python's standard library and avoids the use of sub-processes or shell command execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted intelligence records from public databases.
  • Ingestion points: External API responses from NIST, CISA, and GitHub are processed within scripts/collect_vulnerabilities.py.
  • Boundary markers: The data is normalized into a strictly defined JSON structure as specified by the vulnerability-record.schema.yml and report-contract.md references.
  • Capability inventory: The skill's capabilities are limited to network GET requests to trusted domains and local file writes to user-specified output paths.
  • Sanitization: Data is serialized using standard JSON libraries, preventing control-character injection into the structured output.
  • [SAFE]: The skill implements defensive programming for local data handling, using tempfile for atomic file updates and os.fchmod to ensure that intelligence reports are created with restricted file permissions (read/write access for the owner only).
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 01:41 PM
Security Audit — agent-trust-hub — soia-safe-track-vulnerability-intel