Skills
skills/matteocervelli/llms/analysis/Gen Agent Trust Hub

analysis

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • COMMAND_EXECUTION (MEDIUM): The skill executes a local Python script scripts/analyze_deps.py via the Bash tool, using a --feature parameter derived from external GitHub issue content. This pattern allows for command injection if the input contains shell metacharacters.
  • PROMPT_INJECTION (LOW): The skill facilitates indirect prompt injection by ingesting untrusted data from GitHub issues and comments to guide the analysis process. Evidence Chain: 1. Ingestion points: mcp__github-mcp__get_issue and mcp__github-mcp__get_issue_comments in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Bash tool and file read/write operations. 4. Sanitization: Absent; the skill lacks explicit instructions to sanitize or escape input from GitHub.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 06:14 PM