ai4pkm-helper
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on the
ai4pkmCLI tool to perform all actions, including orchestrator management, agent execution, and status monitoring. - [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection through its data ingestion capabilities.
- Ingestion points: Processes untrusted content from local directories, Apple Notes folders, and Apple Photos albums.
- Boundary markers: There are no explicit delimiters or instructions provided in the configuration to separate processed content from instructions.
- Capability inventory: The system can execute commands through AI executors like
claude_codeandgemini_cliand manage local filesystem tasks. - Sanitization: The skill lacks evidence of input sanitization or filtering to prevent malicious instructions within ingested data from influencing agent behavior.
Audit Metadata