agents-md
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands to discover project files and manage symbolic links for tool compatibility. * Evidence: find command used for discovery in SKILL.md and references/examples.md. * Evidence: ls command used for state verification in SKILL.md. * Evidence: ln command used to create symbolic links in references/examples.md and references/templates.md.
- [PROMPT_INJECTION]: The skill ingests content from existing repository files, creating a potential surface for indirect prompt injection where instructions in those files could influence the agent's behavior. * Ingestion points: Reads existing AGENTS.md and CLAUDE.md files during auditing and migration workflows in SKILL.md. * Boundary markers: Content is read and summarized without explicit delimiters or instructions to ignore embedded directives. * Capability inventory: The skill has file-write capabilities (AGENTS.md) and shell execution capabilities (find, ls, ln). * Sanitization: No sanitization or validation of the ingested file content is performed before the agent processes or summarizes it.
Audit Metadata