recipe-update-doc
Pass
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a basic shell command (
ls) to list existing documentation files in specific project directories (docs/design/,docs/prd/,docs/adr/). This operation is read-only and restricted to non-sensitive paths, posing no security risk.\n- [PROMPT_INJECTION]: The skill contains clear, workflow-oriented instructions. No attempts to override safety guidelines, bypass model constraints, or extract system prompts were detected. The use of terms like 'CRITICAL' refers to document quality gates rather than malicious instruction overrides.\n- [DATA_EXFILTRATION]: No network activity, external URL references, or access to sensitive local files (such as credentials, SSH keys, or environment variables) was found. The skill exclusively processes project documentation.\n- [REMOTE_CODE_EXECUTION]: The skill does not download or execute scripts from external sources. It utilizes a delegation pattern to internal sub-agents (technical-designer,prd-creator,document-reviewer,design-sync) for specialized tasks.\n- [EXTERNAL_DOWNLOADS]: No external packages or remote dependencies are requested or installed by this skill.
Audit Metadata