ticket-autopilot

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs extensive system operations by executing local CLI tools, including Git, the GitHub CLI (gh), and the Azure CLI (az). These operations are centralized within a dedicated command_capture module that implements process containment using Windows Job Objects and POSIX process groups, enforcing strict timeouts and output size limits to prevent resource exhaustion.
  • [INDIRECT_PROMPT_INJECTION]: The system ingests and processes user-provided ticket files from the docs/tickets/ directory. While this data is untrusted, the skill employs strict schema validation and YAML frontmatter parsing to separate metadata from task content. Furthermore, implementation tasks are isolated within independent Git worktrees, minimizing the risk of unauthorized influence on the agent's core scheduling logic.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the synchronization of local agent installations and the bootstrapping of private repositories. These features involve interactions with external registries and the GitHub API. Security is maintained through a requirement for explicit human-authorized evidence and actor tracking for every administrative mutation.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 11:39 AM