herdr-operations
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates shell command execution using the
herdr pane runutility.\n- [INDIRECT_PROMPT_INJECTION]: The skill features a surface for indirect prompt injection by reading potentially untrusted terminal output.\n - Ingestion points: Data is read from terminal logs and agent messages using
herdr agent readandherdr pane wait-output(SKILL.md).\n - Boundary markers: The skill lacks delimiters to separate untrusted data from the agent's instructions.\n
- Capability inventory: The agent can execute shell commands and manipulate the terminal environment (SKILL.md).\n
- Sanitization: Terminal output is processed without sanitization or instruction to ignore embedded commands.
Audit Metadata