threadwave-error-support
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill proactively addresses the risk of indirect prompt injection by defining a mandatory contract that treats all external content from GitHub as untrusted. It specifically instructs the agent to ignore any embedded instructions or commands found within issue bodies or comments, focusing only on verified resolution sections.
- [DATA_EXFILTRATION]: Data privacy is maintained through a strict handoff protocol that filters out sensitive information such as authentication tokens, cookies, and private file paths. The skill further ensures that issue reports are presented to the user in a Markdown block for manual copy-pasting, explicitly stating that no data is automatically sent to external servers.
- [EXTERNAL_DOWNLOADS]: The skill correctly limits external references to the vendor's own verified infrastructure on GitHub and the official product domain for setup guidance and versioning information.
Audit Metadata