skills/xoai/sage/pack-draft/Gen Agent Trust Hub

pack-draft

Pass

Audited by Gen Agent Trust Hub on May 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands such as bash .sage/tools/sage-new-pack.sh <pack-name> and mkdir -p .sage/packs/<pack-name> where <pack-name> is derived from untrusted input data. This creates a risk of command injection if the input contains shell metacharacters.\n- [PROMPT_INJECTION]: The skill exhibits a significant surface for Indirect Prompt Injection by processing untrusted reports to generate prompt instructions.\n
  • Ingestion points: Untrusted data enters via observation-report, processed-sources, and pack-brief specified in the skill frontmatter.\n
  • Boundary markers: Absent. There are no instructions or delimiters to help the agent distinguish between its core instructions and potentially malicious content within the reports.\n
  • Capability inventory: The agent has the ability to execute bash scripts, create directories, and perform extensive file system writes to generate the pack content.\n
  • Sanitization: Absent. The skill does not provide any mechanisms to validate, escape, or sanitize inputs before they are interpolated into shell commands or drafted into new markdown files.
Audit Metadata
Risk Level
SAFE
Analyzed
May 7, 2026, 01:31 AM
Security Audit — agent-trust-hub — pack-draft