herdr
Pass
Audited by Gen Agent Trust Hub on Oct 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on executing the
herdrcommand-line utility to perform operations like creating terminal panes, workspaces, and running background processes. - [INDIRECT_PROMPT_INJECTION]: The skill interacts with external command output and other agents, which presents a surface for indirect prompt injection.
- Ingestion points: The agent reads data from terminal panes via
herdr pane readandherdr agent readin SKILL.md. - Boundary markers: The instructions do not specify the use of delimiters when processing output to prevent embedded instructions from being interpreted as commands.
- Capability inventory: The agent has the ability to execute commands in other panes (
herdr pane run), send keys (herdr agent send-keys), and prompt sub-agents (herdr agent prompt), as documented in SKILL.md. - Sanitization: No specific sanitization methods, filtering, or escaping are mentioned for processing ingested terminal output.
Audit Metadata