skills/agent-sh/agentsys/deslop/Gen Agent Trust Hub

deslop

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local detection script (../../scripts/detect.js) using Node.js to perform the core analysis task.
  • [COMMAND_EXECUTION]: Shell commands including git diff, git symbolic-ref, and xargs are used to determine which files have changed and need scanning.
  • [SAFE]: The skill dynamically loads a local library (../../lib/repo-map) via require for advanced AST-based analysis of the codebase.
  • [PROMPT_INJECTION]: As a tool that ingests and processes content from an external codebase, it presents an indirect prompt injection surface. 1. Ingestion points: Scans local codebase files specified by user or git diff (SKILL.md). 2. Boundary markers: Absent; no markers or instructions to ignore embedded code instructions. 3. Capability inventory: Command execution via node and git (SKILL.md). 4. Sanitization: Absent; no filtering or escaping of analyzed code content.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 07:31 PM