framework-config

Pass

Audited by Gen Agent Trust Hub on Nov 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local Python and Node.js utility scripts located in the .claude/scripts/ directory to facilitate help searches and configuration management. These executions are a core part of the skill's functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to read other skill definition files (SKILL.md) in full. This represents a potential surface for indirect prompt injection if an adversary can place a malicious skill file in the project's .agents/skills/ directory.
  • Ingestion points: Reads files from .agents/skills//SKILL.md.
  • Boundary markers: No explicit delimiters or warnings are mentioned for the content of these external files.
  • Capability inventory: The skill can execute local scripts (.claude/scripts/*) and write to configuration files in the project and home directories.
  • Sanitization: The skill provides strong instructions to the agent to avoid interpolating user-provided text into shell commands and uses strict schema validation (project-config-schema.cjs, ck-config-schema.cjs) for all configuration merges.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 5, 2026, 06:42 PM