Skills
skills/agarichan/sefirot/sefirot-loop/Gen Agent Trust Hub

sefirot-loop

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes sefirot CLI commands, pip, and git commands (merge, add, commit) to manage the development lifecycle.
  • [EXTERNAL_DOWNLOADS]: Recommends installing the sefirot library via pip, which is the core dependency for functionality.
  • [PROMPT_INJECTION]: Surface for indirect prompt injection exists via untrusted data ingestion from milestones.json and design documents (Ingestion points: milestones.json, docs/tasks/; Boundary markers: Specific markdown headers; Capability inventory: Shell and file operations; Sanitization: User review process).
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 10:29 AM