antithesis-query-logs
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from the Antithesis Logs Explorer, which may contain adversarial input from the system under test.\n
- Ingestion points: Text content is extracted from result rows in the
.event_search_resultscontainer byassets/antithesis-query-logs.js.\n - Boundary markers: No specific delimiters or "ignore instructions" warnings are used in the data extraction process.\n
- Capability inventory: The skill uses
agent-browserfor DOM interaction andpython3for processing query parameters.\n - Sanitization: Basic whitespace normalization is applied, but the skill lacks sophisticated sanitization of ingested log content.\n- [EXTERNAL_DOWNLOADS]: The skill references prerequisites and external tools from trusted sources.\n
- Evidence: References to
snouty(from the author's GitHub) andagent-browser(from the Vercel Labs GitHub).
Audit Metadata