skills/simota/agent-skills/orbit/Gen Agent Trust Hub

orbit

Pass

Audited by Gen Agent Trust Hub on Oct 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill's primary function is to generate and execute shell scripts (run-loop.sh, verify.sh, bootstrap.sh, recover.sh, notify.sh) to manage autonomous agent loops.
  • Evidence: Script templates are defined in reference/script-template-runner.md and reference/script-template-support.md.
  • [DYNAMIC_EXECUTION]: The generated runner uses bash -c to execute a configurable EXEC_CMD and a portable timeout utility implemented via perl -e code.
  • Evidence: The run_with_budget and portable_timeout functions in reference/script-template-runner.md.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted agent output from tool calls and files like done.md to generate git commit messages and update loop progress, creating a potential injection surface.
  • Evidence: The squash_iteration_branch function in reference/script-template-runner.md captures LLM output for use in git commit commands.
  • Mitigation: The skill implements robust defenses including SHA-256 pinning of goal files (GOAL_IMMUTABLE), verification scripts (ARTIFACT_PINS), and agent settings (.claude/settings.json), alongside a triple-gate verification process to ensure tasks are completed correctly and without malicious stubs.
  • [EXTERNAL_DOWNLOADS]: The notify.sh template references the edge-tts utility for text-to-speech notifications.
  • Evidence: Found in the notify.sh template within reference/script-template-support.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 28, 2026, 01:21 PM