013-agile-feature
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute the
dateterminal command to retrieve the current system time for documentation headers. This is a legitimate and low-risk use of command execution to ensure accurate metadata in generated files. - [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes untrusted external data (epic files or pasted text) to generate documentation.
- Ingestion points: Untrusted data enters the agent context through file reading or user-pasted epic content as described in
references/013-agile-feature.md. - Boundary markers: The instructions do not define explicit delimiters or instructions to ignore embedded commands within the epic content.
- Capability inventory: The agent has the ability to execute terminal commands (
date) and search/read local files (file_search,read_file). - Sanitization: There is no evidence of filtering or sanitization of the external epic content before it is summarized or processed into new feature files.
Audit Metadata