skill-adapter
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves ingesting instructions and logic from potentially untrusted external sources (other plugins).
- Ingestion points: The skill instructions specify searching and reading content from
plugins/community/,plugins/packages/, andplugins/examples/. It specifically targetsplugin.jsondescriptions,commands/*.md,agents/*.md,skills/*/SKILL.md, andscripts/. - Boundary markers: No boundary markers or instructions to ignore or sanitize embedded instructions within the sourced plugins are defined.
- Capability inventory: The skill is granted
Bash(cmd:*)access, providing a high-privilege execution environment for synthesized patterns. - Sanitization: The skill lacks mechanisms to validate the safety of extracted methodologies before applying them to the current task.
- [COMMAND_EXECUTION]: The skill manifest permits unrestricted Bash command execution (
Bash(cmd:*)). While intended for applying adapted methodologies, this provides a powerful vector for executing commands derived from the ingestion of third-party plugin content.
Audit Metadata