write-prompt-guide

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI tool (gh) to fetch information about open issues and pull requests (gh issue list, gh pr list). This is a legitimate use of the tool to gather information about a repository's current status and limitations for documentation purposes.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data by reading GitHub issue descriptions and pull request metadata to identify known bugs or limitations.
  • Ingestion points: Data enters the system via the gh command-line tool, which retrieves content from GitHub repositories.
  • Boundary markers: None explicitly defined for the ingestion of issue text.
  • Capability inventory: The skill's primary function is text processing and documentation generation; it does not contain capabilities for arbitrary code execution or file-system modifications based on the ingested issue content.
  • Sanitization: The skill is instructed to extract facts rather than execute instructions, which serves as a natural barrier to execution-based prompt injection attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 05:53 AM
Security Audit — agent-trust-hub — write-prompt-guide