jira-ticket

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill demonstrates a vulnerability to indirect prompt injection by processing untrusted data from external sources to influence agent actions.
  • Ingestion points: The agent fetches summary, description, comments, and linked Confluence material from Jira issues (SKILL.md).
  • Boundary markers: The instructions lack explicit delimitation or 'ignore instructions' warnings for the fetched external content.
  • Capability inventory: The skill possesses file write and command execution capabilities by delegating implementation to language-specific runtimes via the $dev skill.
  • Sanitization: There is no mention of sanitizing or validating the external text before it is used for implementation planning and autonomous execution.
  • [COMMAND_EXECUTION]: The implementation workflow relies on delegating tasks to tools like ruby-dev or rust-dev, which execute commands and scripts locally based on plans derived from the untrusted Jira data.
  • [SAFE]: The skill includes repository verification logic, such as checking for AGENTS.md and repository-specific markers, which helps ensure the agent operates only within the intended workspace.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 12:28 PM
Security Audit — agent-trust-hub — jira-ticket