architect-specify

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: Detailed analysis of the skill's instructions, templates, and associated scripts indicates no malicious patterns or security risks. The skill operates locally to support architectural documentation workflows.
  • [COMMAND_EXECUTION]: The skill utilizes local shell scripts (scripts/bash/setup-architect.sh and scripts/powershell/setup-architect.ps1) to perform automated discovery of a project's technology stack. These scripts use standard commands (such as grep, find, and ls) to inspect project configuration files in the current repository root.
  • [DATA_EXPOSURE]: The skill scans for common configuration files such as package.json, requirements.txt, and Dockerfile. This data is used solely to initialize architectural documentation and is not transmitted externally or stored in insecure locations.
  • [PROMPT_INJECTION]: The skill ingests user-supplied Product Requirements Documents (PRDs) via $ARGUMENTS. While this is an ingestion point for external content, the risk of indirect prompt injection is mitigated by the skill's focus on mapping input to structured markdown templates (MADR). No instructions intended to bypass safety filters were detected in the skill definition.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 03:14 PM
Security Audit — agent-trust-hub — architect-specify