loop-on-ci

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (gh) to monitor PR checks and retrieve CI logs. The commands gh pr view, gh pr checks, and gh run view are core to its functionality for identifying and resolving build failures. These are well-known tools from a trusted service used here for their intended purpose.
  • [PROMPT_INJECTION]: The skill processes external data via CI logs, creating a surface for indirect prompt injection.
  • Ingestion points: CI logs fetched through gh run view <run-id> --log-failed as specified in the workflow.
  • Boundary markers: The instructions do not define specific delimiters or guardrails to prevent the agent from obeying instructions potentially embedded within the log output.
  • Capability inventory: The agent is expected to fix failures, which involves code modification and repository updates.
  • Sanitization: There is no evidence of sanitization or content validation for the logs being processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 11:52 AM
Security Audit — agent-trust-hub — loop-on-ci