doca-debug
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze external data sources such as program logs and system diagnostics, which represents a potential surface for indirect prompt injection.
- Ingestion points: The agent is directed in
TASKS.md(test step 3) to capture and interpretstdout,stderr,dmesgoutput, andjournalctllogs. - Boundary markers: The instructions do not explicitly provide delimiters or specific formatting to isolate untrusted log content from the agent's instructions.
- Capability inventory: The skill's capabilities are focused on read-only observability and build-time troubleshooting using standard utilities like
pkg-config,doca_caps, andldd. - Sanitization: No explicit sanitization or filtering steps for the ingested log data are described.
Audit Metadata