rail-align
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes project-specific documentation files (e.g., domain.md, CONTEXT.md) to guide its alignment process, which creates a surface for instructions embedded in those files to influence agent behavior.
- Ingestion points: The agent is instructed to read various files from the docs/monorail/ directory.
- Capability inventory: The skill can write files, create directories, and execute directory listing commands (ls).
- Boundary markers: The instructions do not define specific delimiters for separating ingested content from system instructions.
- Sanitization: No explicit content filtering or sanitization is described.
- [DYNAMIC_EXECUTION]: The 'prototype' ticket resolution type in map-mode.md instructs the agent to create 'throwaway runnable' logic harnesses or UI variants to test assumptions.
- [COMMAND_EXECUTION]: The skill utilizes the shell (specifically the 'ls' command) to verify the presence of documentation files and to manage directory naming sequences based on file counts.
Audit Metadata