openspec-archiving-cn

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses standard shell commands including grep, sed, cat, mv, and git to manage local project documentation. These tools are used for their intended purpose of finding text blocks, replacing content, and moving directories within the spec/ folder structure. The use of quoted heredocs ('EOF') in the examples demonstrates a best practice for preventing accidental shell expansion during file writing.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external change files located in spec/changes/ to update resident specification documents. While this creates a surface for processing untrusted content, the skill explicitly provides instructions for manual verification and human-in-the-loop review, and it recommends manual editing in a text editor for complex modifications to ensure safety and accuracy.
  • [DATA_EXFILTRATION]: There are no network operations or access to sensitive system paths (such as SSH keys or environment secrets) detected. All operations are confined to the local project's specification directories.
  • [OBFUSCATION]: No obfuscation techniques, hidden Unicode characters, or encoded payloads were found. The instructions and scripts are provided in clear text.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 06:03 PM