delegation
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface where untrusted data could influence agent behavior through subagent delegation. • Ingestion points: Untrusted task descriptions passed to delegation functions and data read from the
.local/session_plan.mdfile. • Boundary markers: No specific boundary markers or instructions to ignore embedded commands are documented for the subagent's execution context. • Capability inventory: Subagents are explicitly granted powerful capabilities including 'File operations (read, write, edit, glob, grep)', 'Bash commands', and access to other 'Skills' that may contain sensitive integration details like authentication or payments. • Sanitization: The skill does not describe any sanitization or validation logic for external task inputs before they are processed by a subagent. - [COMMAND_EXECUTION]: Subagents are empowered to execute arbitrary shell commands as part of their 'Bash commands' capability. This represents a significant capability that could be abused if the subagent's instructions are compromised by malicious external data.
Audit Metadata