Skills
skills/armagnac/agent-skills/superplan/Gen Agent Trust Hub

superplan

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is designed for task planning and implementation with no detected malicious patterns, obfuscation, or unauthorized access.
  • [COMMAND_EXECUTION]: Uses a restricted set of tools (git, ls, cat, Write, Edit) to manage code and documentation as intended.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its data ingestion capabilities. 1. Ingestion points: External ticket details from Linear, GitHub, and Jira (SKILL.md). 2. Boundary markers: None specified for ticket content. 3. Capability inventory: File system modification (Write, Edit) and restricted shell commands (git, ls, cat) in SKILL.md. 4. Sanitization: No sanitization of external input is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 07:57 AM