super-pm
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the Bash tool to run local scripts such as check-update.sh for version monitoring via Git and pm-selfcheck for health audits. It also suggests command-line execution for search domain configurations.
- [DYNAMIC_EXECUTION]: Includes shell scripts (create-v2-skill.sh) and templates (subagent-skill-template.md) that dynamically generate or modify AI agent skill files on the local filesystem.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the internet via WebSearch and Exa MCP tools. Ingestion points: external search results and market data. Boundary markers: limited or absent in prompt templates. Capability inventory: includes file writing, shell access, and sub-agent spawning. Sanitization: no explicit sanitization of external data is described.
- [EXTERNAL_DOWNLOADS]: Recommends the installation of secondary skills (AnySearch) and MCP servers (Exa MCP) from external GitHub repositories to extend functionality. These are attributed to the same author or well-known service providers.
Audit Metadata