agents-analyze

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection vulnerability surface detected.
  • Ingestion points: The skill systematically scans and reads content from all files matching */skills/*/skill.md, */skills/*/SKILL.md, and agents-plugin/agents/*.md during Step 1.
  • Boundary markers: Absent. The prompt does not specify delimiters or instructions to ignore or isolate instructions found within the analyzed files.
  • Capability inventory: The skill uses Read, Grep, Glob, Bash(ls *), Bash(wc *), and TodoWrite. Crucially, Step 5 allows for the creation of new agent files via TodoWrite based on the analysis of untrusted files.
  • Sanitization: Absent. There is no logic to sanitize or escape the content of the analyzed files before they are processed or used to generate new agent recommendations.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 07:26 PM