mitm-find-insecure
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from network traffic logs (log.txt), creating an attack surface where malicious instructions embedded in web traffic could influence the agent's behavior.
- Ingestion points: Reads log.txt, which is intended to be populated with raw HTTP/HTTPS traffic captured from external sources.
- Boundary markers: No boundary markers or delimiters are defined in the skill instructions to separate the untrusted log content from the agent's core instructions.
- Capability inventory: The agent is tasked with parsing, interpreting, and summarizing the log content to find vulnerabilities, which involves deep processing of potentially adversarial text.
- Sanitization: No sanitization or filtering of the log content is performed or requested before the agent processes the file.
- [COMMAND_EXECUTION]: The skill documentation provides a shell command to capture network traffic using an external tool.
- Evidence: Provides the command
mitmdump --set flow_detail=3 2>&1 | tee log.txtto generate the necessary input data.
Audit Metadata