skills/jmerta/codex-skills/ci-fix/Gen Agent Trust Hub

ci-fix

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION] (LOW): The skill is vulnerable to Indirect Prompt Injection (Category 8) because it ingests and processes data from external, potentially untrusted sources.
  • Ingestion points: The agent reads CI logs via gh run view --log and downloads file artifacts using gh run download.
  • Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat log/artifact content as untrusted data.
  • Capability inventory: The skill has the capability to modify the local filesystem (including .github/workflows/ files) and execute GitHub CLI commands that affect remote repository state.
  • Sanitization: No sanitization or validation of the ingested log content is performed before the agent interprets it to diagnose issues.
  • [COMMAND_EXECUTION] (SAFE): The skill relies heavily on the GitHub CLI (gh). While this involves executing shell commands, these are standard operations for the skill's stated purpose of CI management. The instructions do not involve piping remote content to a shell or executing arbitrary user-provided strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:13 PM