spec-creator

Pass

Audited by Gen Agent Trust Hub on Aug 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions and reference documentation follow a structured logical workflow for principal-engineer-level technical writing. No malicious patterns, obfuscation, or unauthorized data access attempts were detected during the analysis.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an inherent attack surface for indirect prompt injection as it is designed to ingest and process arbitrary user-provided text (brain-dumps, descriptions) to generate specifications. However, this is consistent with the skill's primary purpose, and there are no high-risk capabilities (like network access or command execution) that would make this surface exploitable for malicious gain.
  • Ingestion points: User-provided system descriptions and intake brain-dumps (SKILL.md, Phase 1).
  • Boundary markers: Absent; the skill relies on LLM reasoning to parse input into structured brainstorming domains.
  • Capability inventory: Markdown file creation and writing to /mnt/user-data/outputs/ (SKILL.md, Phase 3).
  • Sanitization: Absent for user input data.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 29, 2026, 06:18 PM
Security Audit — agent-trust-hub — spec-creator