suanli-deployment

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides a bash script (scripts/call.sh) that executes curl to interact with the API. The script is capable of reading the contents of local files when they are specified with a '@' prefix in the command arguments.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources, specifically container logs and pod events, which are not under the direct control of the skill and could be manipulated by third parties to influence agent behavior.
  • Ingestion points: Data retrieved from API endpoints such as /deployment/task/point_log and /deployment/task/pod_event.
  • Boundary markers: The skill's instructions do not include specific boundary markers or delimiters to protect against instructions embedded within retrieved logs.
  • Capability inventory: The skill utilizes a shell script to perform network operations and read local file contents.
  • Sanitization: No sanitization or validation logic is specified for the data returned from the logging and event endpoints before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 11:14 AM