team-collaboration-issue

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill is designed to fetch and analyze content from GitHub issues using user-provided arguments. This external data is untrusted and could contain malicious instructions designed to manipulate the agent's behavior during the resolution process.\n
  • Ingestion points: Data enters the context via gh issue view $ISSUE_NUMBER --comments and other GitHub CLI commands that retrieve issue metadata and body content from the external GitHub repository.\n
  • Boundary markers: The instructions do not define clear delimiters or specific warnings to the agent to disregard or escape instructions that might be embedded within the retrieved issue data.\n
  • Capability inventory: The skill uses powerful CLI tools including gh (GitHub CLI), git, and search utilities like rg and grep. It also instructs the agent to implement code fixes and create pull requests, which involves writing to the local filesystem and interacting with the remote repository.\n
  • Sanitization: There is no evidence of sanitization or validation of the retrieved issue content before it is interpolated into the agent's reasoning and action steps.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 07:08 AM
Security Audit — agent-trust-hub — team-collaboration-issue