gh-actions-build-trigger

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes shell commands including curl for GitHub API interactions and node -e for parsing JSON responses.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes GitHub Action job logs, which are external data sources that could be influenced by an attacker to include malicious instructions.
  • Ingestion points: The skill downloads job logs to /tmp/job.log using the GitHub Actions API in SKILL.md.
  • Boundary markers: There are no explicit boundary markers or instructions to ignore embedded commands when the agent reads or greps the log file.
  • Capability inventory: The skill uses curl, node, and grep, providing a surface for the agent to act on instructions found within the logs.
  • Sanitization: No sanitization or filtering is applied to the log content before it enters the agent's context.
  • [CREDENTIALS_SAFE]: The skill correctly handles sensitive credentials by instructing the user to export the GitHub token as an environment variable (GH_TOKEN) rather than hardcoding it into scripts or files.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 08:25 AM
Security Audit — agent-trust-hub — gh-actions-build-trigger