Skills
Navigate
skills/agentjido/skills/kata-ex-quality/Gen Agent Trust Hub

kata-ex-quality

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/coverage.exs executes git log to identify recently modified files for targeted coverage analysis.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted project data, including Elixir source code and CSV quality logs, which could contain malicious instructions. 1. Ingestion points: Project source code files and quality CSV files located in docs/quality/. 2. Boundary markers: SKILL.md contains explicit instructions to treat source comments and CSV content as untrusted data rather than instructions. 3. Capability inventory: The skill performs local file system writes and executes shell commands for analysis. 4. Sanitization: The scripts/quality_backlog.exs script implements structural validation for CSV data, enforcing schema versions and field constraints.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 01:19 PM