dx-devops-work-item-manage

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the sf CLI, git, and jq to perform project management and version control tasks. This includes listing and creating work items, checking out branches, and pushing code to remote repositories. These operations are within the scope of the developer tools provided by the vendor.
  • [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection (Category 8) because it ingests untrusted metadata from the Salesforce DevOps Center API and processes it within the agent's context.
  • Ingestion points: Work item subject and description fields are retrieved using sf devops work-item list --json (documented in SKILL.md).
  • Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when reading these fields, allowing potential malicious instructions in work items to influence the agent.
  • Capability inventory: The agent possesses capabilities for shell command execution (sf, git), filesystem modification (git commit), and network operations (git push, sf devops review create).
  • Sanitization: There is no explicit sanitization or validation of the text retrieved from the work item fields before it is used in the prompt or passed as arguments to other commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 06:26 AM
Security Audit — agent-trust-hub — dx-devops-work-item-manage