skyrim-runtime-log-forensics
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted runtime logs, which constitutes a surface for indirect prompt injection if malicious instructions are embedded in the logs.
- Ingestion points: The script
scripts/analyze_runtime_logs.pyreads and parses log files provided by the user from the local filesystem. - Boundary markers: No explicit boundary markers or 'ignore' instructions are used when reading the log content.
- Capability inventory: The skill and its associated script are restricted to reading files and performing string/regex parsing. There are no capabilities for network exfiltration, subprocess execution (exec/eval), or file system writes.
- Sanitization: The diagnostic script uses targeted regular expressions to identify specific modding-related failure patterns, which inherently filters out content that does not match the expected log schema.
Audit Metadata