analyze-frame
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
- [COMMAND_EXECUTION] (HIGH): The file
templates/runbook/execute.mddirects the agent or user to execute a Python script located at~/.claude/skills/analyze-frame/scripts/validate_spec.py. Hardcoded paths to scripts in the user's home directory outside the skill's own package are a significant security risk, as the script's integrity and behavior are unverified. - [REMOTE_CODE_EXECUTION] (HIGH): The core functionality of this skill involves an AI agent generating TypeScript code based on external YAML specifications (e.g.,
aggregate.yaml,use-case.yaml). This creates a substantial Indirect Prompt Injection (Category 8) surface; an attacker-provided specification could include instructions designed to influence the code generation process, leading to the creation of backdoored code or exploitation of the agent's capabilities.
Recommendations
- AI detected serious security threats
Audit Metadata