packet-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes network packet captures (.pcap/.pcapng) which originate from external and untrusted sources.
- Ingestion points: The skill ingests data via Slack file uploads through the
save_pcap_from_base64tool and reads existing captures from the local filesystem for analysis. - Boundary markers: The skill instructions do not include boundary markers or explicit prompts to the agent to treat the analysis results as data rather than instructions.
- Capability inventory: The skill allows writing files to the local filesystem and executes a Python-based server script for traffic analysis.
- Sanitization: There is no evidence of sanitization or filtering applied to packet data (e.g., DNS queries, HTTP headers) before it is summarized by the agent.
Audit Metadata