action-controller
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The instructions direct the agent to read local project files and command outputs to ensure consistency with the existing codebase. If these files are manipulated by an attacker, they could contain instructions that influence the agent's behavior.
- Ingestion points: The skill uses 'cat' to read 'app/controllers/application_controller.rb' and 'bin/rails routes' to inspect the project state (documented in SKILL.md).
- Boundary markers: The skill lacks explicit instructions to treat the content of existing files as untrusted data or to ignore any agent-targeted instructions found within them.
- Capability inventory: The skill possesses 'Write', 'Edit', and 'Bash' capabilities (limited to Rails generators and routes), which could be leveraged to modify the application based on injected instructions.
- Sanitization: There is no evidence of filtering or sanitization applied to the data ingested from the project's local environment.
Audit Metadata