gen-specs-as-issues
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is entirely composed of instructional markdown and does not contain any executable scripts or binary files.
- [PROMPT_INJECTION]: No instructions were found that attempt to override AI safety guardrails, extract system prompts, or bypass behavioral constraints.
- [DATA_EXFILTRATION]: The skill does not include commands to access sensitive local files, hardcode credentials, or transmit data to unauthorized external domains.
- [REMOTE_CODE_EXECUTION]: There are no patterns involving the download or execution of remote scripts or unverifiable software packages.
- [INDIRECT_PROMPT_INJECTION]: While the skill involves reading project documentation which is an untrusted data source, it provides a safe, structured process for human-in-the-loop issue creation without automated execution of embedded commands.
Audit Metadata