check-ci

Pass

Audited by Gen Agent Trust Hub on Mar 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local shell script scripts/check_ci.sh which interacts with the gh (GitHub CLI) tool. The script correctly handles variables with double-quoting (e.g., "$PR", "$OWNER/$REPO") to prevent command injection from user-supplied parameters.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it instructs the agent to parse CI logs and check names which contain untrusted data from the CI environment.
  • Ingestion points: The agent is directed to read output from gh pr checks and workflow logs via gh run view --log-failed (SKILL.md).
  • Boundary markers: No delimiters or isolation markers are defined to separate CI log content from agent instructions.
  • Capability inventory: The agent has the capability to execute development tools and shell commands (npm, pytest, ruff, etc.) to resolve failures.
  • Sanitization: No sanitization or filtering is applied to the CI log data before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 24, 2026, 09:57 PM