suanli-job

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a local script, scripts/call.sh, to communicate with the service API. This script executes curl to send data and python3 to generate timestamps for request headers.- [EXTERNAL_DOWNLOADS]: The skill interacts with official endpoints including openapi.suanli.cn for orchestration and harbor.suanleme.cn for container images. These are legitimate service domains belonging to the vendor for managing and deploying compute workloads.- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from the platform, such as execution logs and task status reports, which could potentially contain untrusted content from the task environment.
  • Ingestion points: API responses from /api/deployment/task/point_log and /api/task/job/detail processed via the command-line caller.
  • Boundary markers: The skill does not explicitly define delimiters to isolate external log data from the agent's core instructions.
  • Capability inventory: The agent can trigger job creation, updates, and termination via the provided API wrapper scripts based on its findings.
  • Sanitization: No explicit sanitization or filtering of external API data is performed prior to ingestion.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 11:14 AM